Task-Boundary Mode
How task-instance boundaries are drawn from the event stream. Applies to every Task SoP, Step SoP, and Variants view.
Case Advancement
📝 Task SoP
Mined from 1,184 instances
Advance Case
How 15 analysts execute this task across 425 cases — auto-generated Standard Operating Procedure with adherence scoring.
Sub-Task Filter
Reshapes all tabs below to just this anchor within the composite task.
85.4
Avg Adherence Score (0-100)
Median 82.5 · Range 37.2–97.5
15.2 min
Avg Task Duration
0.3 min median
2.5
Avg Sub-Steps per Instance
74.5% have multiple sub-steps
14.6%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
14.6% of instances follow this exactly
Advance Case — Mined SoP
The most common sub-step sequence observed across 2,467 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Complete Action
1.1s
0.6 edits
Veeva Safety
66%
Variant Explorer
Reference
Advance Case Reference Path
360 instances · 11 analysts
This is the mined reference path — 360 instances across 11 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Advance Case.
Genuine Complexity
Complex-Case Execution Path
255 instances · 12 analysts
· Δ adherence +11.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Approval Complete
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 97.4 vs reference 85.4 (delta +11.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
166 instances · 5 analysts
· Δ adherence +11.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Data Entry Complete
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 97.3 vs reference 85.4 (delta +11.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
157 instances · 15 analysts
· Δ adherence -1.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Save Changes
reference at this step: Complete Action
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 83.9 vs reference 85.4 (delta -1.5). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
144 instances · 10 analysts
· Δ adherence -2.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Start Approval Workflow
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 82.7 vs reference 85.4 (delta -2.8). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
141 instances · 7 analysts
· Δ adherence -2.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Continue Workflow
reference at this step: Complete Action
2
Delete Item
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 82.6 vs reference 85.4 (delta -2.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
112 instances · 10 analysts
· Δ adherence -2.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Continue Workflow
reference at this step: Complete Action
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 83.4 vs reference 85.4 (delta -2.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
62 instances · 7 analysts
· Δ adherence +12.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Send to Case Approval
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 97.5 vs reference 85.4 (delta +12.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
60 instances · 3 analysts
· Δ adherence -1.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Open Cases Section
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 83.6 vs reference 85.4 (delta -1.8). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
33 instances · 6 analysts
· Δ adherence -1.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Navigate to Cases
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 83.6 vs reference 85.4 (delta -1.8). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
33 instances · 7 analysts
· Δ adherence +12.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Send for QC
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 97.5 vs reference 85.4 (delta +12.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
31 instances · 10 analysts
· Δ adherence +11.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Select Approval Complete
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 97.3 vs reference 85.4 (delta +11.8). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
30 instances · 9 analysts
· Δ adherence -2.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Select Start Approval Workflow
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 82.5 vs reference 85.4 (delta -2.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 5,258 | 86.8% | 98.8% |
| Microsoft Word | 149 | 2.5% | 0.0% |
| Acrobat | 130 | 2.1% | 0.0% |
| Microsoft Teams | 126 | 2.1% | 0.0% |
| collaboration.merck.com | 86 | 1.4% | 0.0% |
| Microsoft Excel | 79 | 1.3% | 0.0% |
| Microsoft Outlook | 74 | 1.2% | 0.0% |
| usc-excel.officeapps.live.com | 25 | 0.4% | 0.0% |
| gpteal.merck.com | 14 | 0.2% | 0.0% |
| Notepad | 12 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Veeva Safety | Microsoft Teams | 84 |
| Veeva Safety | Acrobat | 74 |
| Veeva Safety | Microsoft Word | 69 |
| Veeva Safety | Microsoft Outlook | 56 |
| Veeva Safety | Microsoft Excel | 50 |
| Acrobat | Microsoft Word | 48 |
| Acrobat | Veeva Safety | 46 |
| Microsoft Teams | Veeva Safety | 46 |
| Microsoft Word | Veeva Safety | 42 |
| Veeva Safety | collaboration.merck.com | 41 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 14,776 events 425 cases 232.6 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
1943×
pyzeClick
<a>
"Show more"
476×
pyzeClick
<span>
"Narrative"
470×
pyzeClick
<span>
"Continue"
447×
pyzeClick
<button>
"Save"
422×
pyzeClick
<li>
"Narrative"
395×
pyzeClick
<span>
"Adverse Events"
349×
pyzeClick
<li>
"Details"
325×
Phobos DOM 2,387 events 289 cases 72.5 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
359×
pyzeClick
<button>
"Advance"
225×
pyzeClick
<button>
"Save Comments"
183×
pyzeClick
<button>
"Send to process"
114×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
102×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
94×
pyzeClick
<a>
"Cases"
85×
pyzeClick
<button>
"Send to Medical Review"
68×
Microsoft Word Document 571 events 149 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 552979 (English) (26689105_0_2_348832)
1638×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
723×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
572×
file
V-Narrative 549550 (English) (26643764_0_1_347593)
526×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
382×
file
V-Narrative 533882 (English) (26397511_0_1_345751)
308×
file
V-Narrative 553023 (English) (26689066_0_2_351253)
257×
file
V-Narrative 542512 (English) (26522671_0_2_344761)
255×
Acrobat App 423 events 98 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
5,505 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 358 events 143 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
4,239 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Excel Document 202 events 81 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Case tracking list
696×
file
Daily tracking list (PV-SMART) copy
435×
file
1761620 (v2.1)
330×
file
Priorización
314×
file
2352822 (v0.1)
234×
file
E2B_MSDPRO_014172_AGPROD1_20251127163042
155×
file
I-0000353124
142×
file
I-0000350382
141×
Microsoft Outlook Message 194 events 108 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
collaboration.merck.com App 176 events 83 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Notepad App 50 events 27 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
ONENOTE Document 48 events 29 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.07
Swivel Rate
events in non-primary app / total
0.66
Zero-Edit Visit Rate
inspection-only clicks
1.08
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.66
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.66
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.08
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.07
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2309351 | Analyst 01 | 37.2 | 15.3m | 25 | 0.84 | 0.00 | 11 |
| 2345589 | Analyst 13 | 42.1 | 2.0m | 14 | 0.93 | 0.00 | 6 |
| 2235982 | Analyst 13 | 45.5 | 1.6m | 10 | 0.90 | 0.00 | 5 |
| 2341558 | Analyst 13 | 47.5 | 11.3m | 24 | 0.67 | 0.00 | 7 |
| 2354204 | Analyst 13 | 48.0 | 1.7m | 10 | 0.90 | 0.00 | 4 |
Users on This Task
15 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
06
15
14
07
04
11
01
03
02
09
05
12
10
16
13
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 06 | 6 | 3.0 | 90.0 | Zero-Edit Visits | 0.00 | 0.00 | 10 |
|
Reference SoP
1
Complete Action
Analyst 06's Dominant Sequence
— 1 of 6 instances
1
Complete Action
This analyst exhibits 6 distinct sequences across their 6 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 25 | 25.0 | 88.3 | Zero-Edit Visits | 0.00 | 0.00 | 27 |
|
Reference SoP
1
Complete Action
Analyst 15's Dominant Sequence
— 6 of 25 instances
1
Complete Action
2
Navigate to Cases
This analyst exhibits 14 distinct sequences across their 25 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 69 | 34.5 | 87.9 | Zero-Edit Visits | 0.00 | 0.00 | 72 |
|
Reference SoP
1
Complete Action
Analyst 14's Dominant Sequence
— 8 of 69 instances
1
Complete Action
2
Start Approval Workflow
This analyst exhibits 36 distinct sequences across their 69 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 17 | 17.0 | 87.5 | Zero-Edit Visits | 0.00 | 0.00 | 14 |
|
Reference SoP
1
Complete Action
Analyst 07's Dominant Sequence
— 3 of 17 instances
1
Save Changes
· reference: Complete Action
2
Preview English Narrative
This analyst exhibits 12 distinct sequences across their 17 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 401 | 28.6 | 87.4 | Zero-Edit Visits | 0.00 | 0.00 | 911 |
|
Reference SoP
1
Complete Action
Analyst 04's Dominant Sequence
— 55 of 401 instances
1
Complete Action
2
Open Cases Section
This analyst exhibits 87 distinct sequences across their 401 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 12 | 12.0 | 86.7 | Zero-Edit Visits | 0.00 | 0.00 | 22 |
|
Reference SoP
1
Complete Action
Analyst 11's Dominant Sequence
— 3 of 12 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 10 distinct sequences across their 12 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 351 | 23.4 | 86.4 | Zero-Edit Visits | 0.07 | 0.00 | 16,975 |
|
Reference SoP
1
Complete Action
Analyst 01's Dominant Sequence
— 77 of 351 instances
1
Complete Action
This analyst exhibits 111 distinct sequences across their 351 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 303 | 23.3 | 85.9 | Zero-Edit Visits | 0.05 | 0.00 | 4,382 |
|
Reference SoP
1
Complete Action
Analyst 03's Dominant Sequence
— 54 of 303 instances
1
Complete Action
This analyst exhibits 90 distinct sequences across their 303 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 353 | 29.4 | 85.6 | Zero-Edit Visits | 0.06 | 0.00 | 3,617 |
|
Reference SoP
1
Complete Action
Analyst 02's Dominant Sequence
— 77 of 353 instances
1
Complete Action
This analyst exhibits 124 distinct sequences across their 353 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 390 | 24.4 | 84.7 | Zero-Edit Visits | 0.08 | 0.00 | 7,612 |
|
Reference SoP
1
Complete Action
Analyst 09's Dominant Sequence
— 65 of 390 instances
1
Continue Workflow
· reference: Complete Action
2
Delete Item
This analyst exhibits 96 distinct sequences across their 390 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 191 | 19.1 | 83.9 | Zero-Edit Visits | 0.11 | 0.00 | 623 |
|
Reference SoP
1
Complete Action
Analyst 05's Dominant Sequence
— 27 of 191 instances
1
Complete Action
This analyst exhibits 89 distinct sequences across their 191 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 208 | 23.1 | 82.7 | Zero-Edit Visits | 0.11 | 0.00 | 1,861 |
|
Reference SoP
1
Complete Action
Analyst 12's Dominant Sequence
— 21 of 208 instances
1
Complete Action
This analyst exhibits 110 distinct sequences across their 208 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 2 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Complete Action
Analyst 10's Dominant Sequence
— 1 of 2 instances
1
Continue Workflow
· reference: Complete Action
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 16 | 2 | 2.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 7 |
|
Reference SoP
1
Complete Action
Analyst 16's Dominant Sequence
— 1 of 2 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 137 | 12.5 | 82.2 | Zero-Edit Visits | 0.16 | 0.00 | 1,338 |
|
Reference SoP
1
Complete Action
Analyst 13's Dominant Sequence
— 26 of 137 instances
1
Complete Action
This analyst exhibits 60 distinct sequences across their 137 instances; the dominant one is shown.
|
||||||||
88.0
Avg Adherence Score (0-100)
Median 82.5 · Range 42.1–97.5
18.7 min
Avg Task Duration
0.2 min median
2.1
Avg Sub-Steps per Instance
78.0% have multiple sub-steps
—
Reference Path Coverage
1,637 sub-task instances
Reference Baseline
Advance Case — Mined SoP
The most common sub-step sequence observed across 1,637 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Complete Action
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 3,175 | 91.0% | 98.3% |
| Microsoft Teams | 65 | 1.9% | 0.0% |
| Microsoft Outlook | 44 | 1.3% | 0.0% |
| Microsoft Excel | 43 | 1.2% | 0.0% |
| collaboration.merck.com | 42 | 1.2% | 0.0% |
| Acrobat | 29 | 0.8% | 0.0% |
| Microsoft Word | 22 | 0.6% | 0.0% |
| usc-excel.officeapps.live.com | 15 | 0.4% | 0.0% |
| gpteal.merck.com | 6 | 0.2% | 0.0% |
| apps.powerapps.com | 4 | 0.1% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Veeva Safety | Microsoft Teams | 46 |
| Veeva Safety | Microsoft Outlook | 34 |
| Veeva Safety | Microsoft Excel | 33 |
| Veeva Safety | Acrobat | 22 |
| Microsoft Teams | Veeva Safety | 21 |
| Microsoft Outlook | Veeva Safety | 19 |
| Veeva Safety | Microsoft Word | 17 |
| Veeva Safety | collaboration.merck.com | 16 |
| collaboration.merck.com | usc-excel.officeapps.live.com | 13 |
| Acrobat | Veeva Safety | 12 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 9,857 events 378 cases 132.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
1941×
pyzeClick
<a>
"Show more"
368×
pyzeClick
<span>
"Narrative"
311×
pyzeClick
<li>
"Narrative"
308×
pyzeClick
<span>
"Approval Complete"
306×
pyzeClick
<span>
"Start Approval Workflow"
252×
pyzeClick
<li>
"Details"
243×
pyzeClick
<span>
"Validation Results (Failures & Warnings)"
199×
Phobos DOM 1,934 events 255 cases 55.5 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
322×
pyzeClick
<button>
"Advance"
197×
pyzeClick
<button>
"Save Comments"
161×
pyzeClick
<button>
"Send to process"
88×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
87×
pyzeClick
<button>
"Send to Medical Review"
67×
pyzeClick
<a>
"Cases"
65×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
63×
Microsoft Word Document 329 events 129 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 528191 (English) (26300681_0_1_338349)
382×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
306×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
294×
file
V-Narrative 542512 (English) (26522671_0_2_344761)
255×
file
V-Narrative 531284 (English) (26355806_0_2_340584)
225×
file
V-Narrative 525091 (English) (26253532_0_1_338428)
201×
file
V-Narrative 531868 (English) (26362311_0_1_340534)
185×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
Microsoft Teams App 217 events 107 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,461 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Acrobat App 214 events 83 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,067 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Outlook Message 127 events 80 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
subj
RE: [Confidential] <Due on 24H> <Approved> <Correction multiple cases>Product Selection
32×
subj
<compensación> Case#2252065
31×
subj
<Due today><Void request><GPVCM POLAND> Case # 2326062
23×
subj
<GPVCM Announcement> Promotions to Safety Data Manager positions in Poland
18×
subj
<compensación> Case#2351734
17×
subj
Due today reports on 03-DEC-2025
14×
subj
Due today reports on 01-DEC-2025
13×
subj
RE: <compensación> Case#2309077
8×
Microsoft Excel Document 105 events 59 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
collaboration.merck.com App 98 events 52 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
ONENOTE Document 41 events 27 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Notepad App 33 events 18 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.05
Swivel Rate
events in non-primary app / total
0.53
Zero-Edit Visit Rate
inspection-only clicks
1.02
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.53
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.53
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.02
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.05
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2345589 | Analyst 13 | 42.1 | 2.0m | 14 | 0.93 | 0.00 | 6 |
| 2235982 | Analyst 13 | 45.5 | 1.6m | 10 | 0.90 | 0.00 | 5 |
| 2350053 | Analyst 02 | 50.1 | 4.5m | 9 | 0.89 | 0.30 | 1 |
| 2350576 | Analyst 13 | 51.0 | 0.4m | 10 | 0.80 | 0.00 | 4 |
| 2291397 | Analyst 13 | 56.0 | 5.4m | 5 | 0.80 | 0.00 | 2 |
Users on This Task
14 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
06
11
07
15
04
14
03
01
02
09
05
12
13
16
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 06 | 3 | 3.0 | 92.5 | Zero-Edit Visits | 0.00 | 0.00 | 8 |
|
Reference SoP
1
Complete Action
Analyst 06's Dominant Sequence
— 1 of 3 instances
1
Complete Action
This analyst exhibits 3 distinct sequences across their 3 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 7 | 7.0 | 89.6 | Zero-Edit Visits | 0.00 | 0.00 | 5 |
|
Reference SoP
1
Complete Action
Analyst 11's Dominant Sequence
— 1 of 7 instances
1
Complete Action
2
Select Send to Case Approval
This analyst exhibits 7 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 8 | 8.0 | 89.4 | Zero-Edit Visits | 0.00 | 0.00 | 5 |
|
Reference SoP
1
Complete Action
Analyst 07's Dominant Sequence
— 2 of 8 instances
1
Complete Action
2
Select Start Approval Workflow
This analyst exhibits 6 distinct sequences across their 8 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 19 | 19.0 | 89.1 | Zero-Edit Visits | 0.00 | 0.00 | 15 |
|
Reference SoP
1
Complete Action
Analyst 15's Dominant Sequence
— 6 of 19 instances
1
Complete Action
2
Navigate to Cases
This analyst exhibits 10 distinct sequences across their 19 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 255 | 18.2 | 89.0 | Zero-Edit Visits | 0.00 | 0.00 | 624 |
|
Reference SoP
1
Complete Action
Analyst 04's Dominant Sequence
— 55 of 255 instances
1
Complete Action
2
Open Cases Section
This analyst exhibits 41 distinct sequences across their 255 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 52 | 26.0 | 89.0 | Zero-Edit Visits | 0.00 | 0.00 | 36 |
|
Reference SoP
1
Complete Action
Analyst 14's Dominant Sequence
— 8 of 52 instances
1
Complete Action
2
Start Approval Workflow
This analyst exhibits 22 distinct sequences across their 52 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 204 | 15.7 | 88.4 | Zero-Edit Visits | 0.04 | 0.00 | 3,047 |
|
Reference SoP
1
Complete Action
Analyst 03's Dominant Sequence
— 54 of 204 instances
1
Complete Action
This analyst exhibits 48 distinct sequences across their 204 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 260 | 17.3 | 88.2 | Zero-Edit Visits | 0.07 | 0.00 | 16,803 |
|
Reference SoP
1
Complete Action
Analyst 01's Dominant Sequence
— 77 of 260 instances
1
Complete Action
This analyst exhibits 58 distinct sequences across their 260 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 240 | 20.0 | 88.1 | Zero-Edit Visits | 0.05 | 0.00 | 202 |
|
Reference SoP
1
Complete Action
Analyst 02's Dominant Sequence
— 77 of 240 instances
1
Complete Action
This analyst exhibits 64 distinct sequences across their 240 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 228 | 14.2 | 88.1 | Zero-Edit Visits | 0.05 | 0.00 | 7,306 |
|
Reference SoP
1
Complete Action
Analyst 09's Dominant Sequence
— 42 of 228 instances
1
Complete Action
2
Mark Approval Complete
This analyst exhibits 36 distinct sequences across their 228 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 144 | 14.4 | 87.2 | Zero-Edit Visits | 0.06 | 0.00 | 439 |
|
Reference SoP
1
Complete Action
Analyst 05's Dominant Sequence
— 27 of 144 instances
1
Complete Action
This analyst exhibits 54 distinct sequences across their 144 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 111 | 15.9 | 86.9 | Zero-Edit Visits | 0.07 | 0.00 | 1,001 |
|
Reference SoP
1
Complete Action
Analyst 12's Dominant Sequence
— 21 of 111 instances
1
Complete Action
This analyst exhibits 45 distinct sequences across their 111 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 105 | 10.5 | 85.1 | Zero-Edit Visits | 0.13 | 0.00 | 1,100 |
|
Reference SoP
1
Complete Action
Analyst 13's Dominant Sequence
— 26 of 105 instances
1
Complete Action
This analyst exhibits 40 distinct sequences across their 105 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 16 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Complete Action
Analyst 16's Dominant Sequence
— 1 of 1 instances
1
Complete Action
|
||||||||
81.0
Avg Adherence Score (0-100)
Median 82.5 · Range 37.2–97.5
12.3 min
Avg Task Duration
0.3 min median
2.9
Avg Sub-Steps per Instance
73.7% have multiple sub-steps
—
Reference Path Coverage
426 sub-task instances
Reference Baseline
Advance Case — Mined SoP
The most common sub-step sequence observed across 426 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Continue Workflow
2
Notepad
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 1,067 | 85.0% | 100.0% |
| Acrobat | 56 | 4.5% | 0.0% |
| Microsoft Word | 54 | 4.3% | 0.0% |
| Microsoft Teams | 24 | 1.9% | 0.0% |
| Microsoft Outlook | 9 | 0.7% | 0.0% |
| collaboration.merck.com | 8 | 0.6% | 0.0% |
| Microsoft Excel | 8 | 0.6% | 0.0% |
| gpteal.merck.com | 6 | 0.5% | 0.0% |
| TextInputHost | 3 | 0.2% | 0.0% |
| YoudaoDict | 2 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Microsoft Word | Acrobat | 27 |
| Acrobat | Microsoft Word | 27 |
| Veeva Safety | Acrobat | 23 |
| Acrobat | Veeva Safety | 19 |
| Veeva Safety | Microsoft Word | 19 |
| Veeva Safety | Microsoft Teams | 18 |
| Microsoft Word | Veeva Safety | 12 |
| Microsoft Teams | Veeva Safety | 8 |
| Veeva Safety | Microsoft Excel | 6 |
| Veeva Safety | Microsoft Outlook | 6 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 4,654 events 167 cases 77.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Continue"
447×
pyzeClick
<span>
"Complete"
327×
pyzeClick
<li>
"Delete"
300×
pyzeClick
<button>
"Create"
151×
pyzeClick
<a>
"Show more"
118×
pyzeClick
<li>
"Narrative"
112×
pyzeClick
<span>
"Narrative"
103×
pyzeClick
<li>
"Details"
92×
Phobos DOM 305 events 57 cases 11.0 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
57×
pyzeClick
<button>
"Advance"
33×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
24×
pyzeClick
<button>
"Save Comments"
18×
pyzeClick
<a>
"Cases"
17×
pyzeClick
<button>
">> Send to Quality Check"
13×
pyzeClick
<a>
"Cases Coordinator"
11×
pyzeClick
<button>
"Send to process"
10×
Microsoft Word Document 189 events 48 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 552979 (English) (26689105_0_2_348832)
557×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
515×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
429×
file
V-Narrative 553023 (English) (26689066_0_2_351253)
251×
file
V-Narrative 533882 (English) (26397511_0_1_345751)
230×
file
V-Narrative 549428 (English) (26642399_0_2_349187)
170×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
file
V-Narrative 538518 (English) (26459147_0_1_344628)
169×
Acrobat App 154 events 38 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,365 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 109 events 50 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,019 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Excel Document 54 events 24 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
1761620 (v2.1)
330×
file
2352822 (v0.1)
216×
file
Case tracking list
168×
file
MERCK2516915_ARGPRO_20251125054618
97×
file
Daily tracking list (PV-SMART) copy
87×
file
Priorización
68×
file
ICHICSR-EVHUMAN_10020432161 (1)
30×
file
GDMS800-039-001 GPVCM Finding Definitions (1)
20×
Microsoft Outlook Message 51 events 34 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
collaboration.merck.com App 36 events 21 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Notepad App 19 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
gpteal.merck.com DOM 18 events 10 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.07
Swivel Rate
events in non-primary app / total
0.92
Zero-Edit Visit Rate
inspection-only clicks
1.19
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.92
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.92
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.19
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.07
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2309351 | Analyst 01 | 37.2 | 15.3m | 25 | 0.84 | 0.00 | 11 |
| 2341558 | Analyst 13 | 47.5 | 11.3m | 24 | 0.67 | 0.00 | 7 |
| 2252065 | Analyst 05 | 49.4 | 5.4m | 16 | 0.69 | 0.00 | 6 |
| 2350507 | Analyst 09 | 50.0 | 13.7m | 12 | 0.75 | 0.00 | 5 |
| 2347303 | Analyst 02 | 55.9 | 5.7m | 11 | 0.64 | 0.00 | 4 |
Users on This Task
14 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
14
04
15
07
11
10
06
02
01
03
09
12
05
13
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 14 | 5 | 5.0 | 86.2 | Zero-Edit Visits | 0.00 | 0.00 | 10 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 14's Dominant Sequence
— 1 of 5 instances
1
Continue Workflow
2
Press Enter
· reference: Notepad
3
Continue Link
4
Select List Item
5
View Study
6
Select Change Type
7
View External Product
This analyst exhibits 5 distinct sequences across their 5 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 66 | 6.0 | 83.7 | Zero-Edit Visits | 0.00 | 0.00 | 177 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 04's Dominant Sequence
— 31 of 66 instances
1
Continue Workflow
This analyst exhibits 24 distinct sequences across their 66 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 2 | 2.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 15's Dominant Sequence
— 1 of 2 instances
1
Continue Workflow
2
View Drug History
· reference: Notepad
3
View Adverse Events
4
View Products
5
View Case Assessments
6
View Case Assessment Results
7
Select Generate Narrative
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 2 | 2.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 07's Dominant Sequence
— 1 of 2 instances
1
Continue Workflow
2
Show More Details
· reference: Notepad
3
Select Reference Numbers
4
Delete Item
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 11's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
2
View Details Menu
· reference: Notepad
|
||||||||
| ▸ | Analyst 10 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 10's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
|
||||||||
| ▸ | Analyst 06 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 06's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
|
||||||||
| ▸ | Analyst 02 | 68 | 6.8 | 81.6 | Zero-Edit Visits | 0.06 | 0.00 | 3,276 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 02's Dominant Sequence
— 32 of 68 instances
1
Continue Workflow
2
Delete Item
· reference: Notepad
This analyst exhibits 27 distinct sequences across their 68 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 52 | 3.7 | 81.4 | Zero-Edit Visits | 0.08 | 0.00 | 104 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 01's Dominant Sequence
— 16 of 52 instances
1
Continue Workflow
This analyst exhibits 25 distinct sequences across their 52 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 51 | 3.9 | 80.9 | Zero-Edit Visits | 0.06 | 0.00 | 1,258 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 03's Dominant Sequence
— 22 of 51 instances
1
Continue Workflow
This analyst exhibits 19 distinct sequences across their 51 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 106 | 8.8 | 80.8 | Zero-Edit Visits | 0.07 | 0.00 | 99 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 09's Dominant Sequence
— 65 of 106 instances
1
Continue Workflow
2
Delete Item
· reference: Notepad
This analyst exhibits 26 distinct sequences across their 106 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 34 | 6.8 | 79.4 | Zero-Edit Visits | 0.13 | 0.00 | 78 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 12's Dominant Sequence
— 8 of 34 instances
1
Continue Workflow
2
Delete Item
· reference: Notepad
This analyst exhibits 24 distinct sequences across their 34 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 20 | 2.9 | 76.8 | Zero-Edit Visits | 0.19 | 0.00 | 70 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 05's Dominant Sequence
— 5 of 20 instances
1
Continue Workflow
This analyst exhibits 14 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 17 | 1.9 | 75.5 | Zero-Edit Visits | 0.22 | 0.00 | 181 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 13's Dominant Sequence
— 8 of 17 instances
1
Continue Workflow
This analyst exhibits 10 distinct sequences across their 17 instances; the dominant one is shown.
|
||||||||
79.8
Avg Adherence Score (0-100)
Median 82.5 · Range 48.0–97.5
4.0 min
Avg Task Duration
0.6 min median
3.3
Avg Sub-Steps per Instance
61.1% have multiple sub-steps
—
Reference Path Coverage
404 sub-task instances
Reference Baseline
Advance Case — Mined SoP
The most common sub-step sequence observed across 404 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Save Changes
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 1,016 | 77.3% | 100.0% |
| Microsoft Word | 73 | 5.6% | 0.0% |
| Acrobat | 45 | 3.4% | 0.0% |
| Microsoft Teams | 37 | 2.8% | 0.0% |
| collaboration.merck.com | 36 | 2.7% | 0.0% |
| Microsoft Excel | 28 | 2.1% | 0.0% |
| Microsoft Outlook | 21 | 1.6% | 0.0% |
| usc-excel.officeapps.live.com | 10 | 0.8% | 0.0% |
| Notepad | 8 | 0.6% | 0.0% |
| evocpr01.apps.americas.cf.merck.com | 5 | 0.4% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Veeva Safety | Microsoft Word | 36 |
| Veeva Safety | Acrobat | 27 |
| Microsoft Word | Veeva Safety | 22 |
| Veeva Safety | Microsoft Teams | 22 |
| Veeva Safety | collaboration.merck.com | 19 |
| Acrobat | Microsoft Word | 17 |
| Microsoft Teams | Veeva Safety | 17 |
| Acrobat | Veeva Safety | 17 |
| Veeva Safety | Microsoft Outlook | 14 |
| Microsoft Word | Acrobat | 14 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 6,667 events 254 cases 125.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
449×
pyzeClick
<button>
"Save"
422×
pyzeClick
<span>
"Narrative"
217×
pyzeClick
<li>
"Narrative"
183×
pyzeClick
<a>
"Show more"
183×
pyzeClick
<span>
"Adverse Events"
179×
pyzeClick
<li>
"Details"
149×
pyzeClick
<button>
"Create"
133×
Phobos DOM 754 events 111 cases 26.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
105×
pyzeClick
<button>
"Advance"
71×
pyzeClick
<button>
"Save Comments"
59×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
52×
pyzeClick
<button>
"Send to process"
40×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
24×
pyzeClick
<a>
"Cases"
22×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
19×
Microsoft Word Document 281 events 81 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 552979 (English) (26689105_0_2_348832)
1076×
file
V-Narrative 549550 (English) (26643764_0_1_347593)
526×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
515×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
377×
file
V-Narrative 533909 (English) (26397745_0_2_338171)
243×
file
V-Narrative 545329 (English) (26567747_0_2_344796)
215×
file
V-Narrative 540358 (English) (26494608_0_2_342069)
195×
file
Narrative 526208 (English)
190×
Acrobat App 207 events 57 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
3,088 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 163 events 68 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,132 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Excel Document 110 events 44 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Case tracking list
382×
file
2352822 (v0.1)
234×
file
Daily tracking list (PV-SMART) copy
210×
file
I-0000353124
142×
file
I-0000350382
141×
file
Priorización
61×
file
MERCK2524493_ARGPRO_20251202045619
46×
file
ICHICSR-EVHUMAN_10020432161 (1)
37×
collaboration.merck.com App 100 events 48 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Microsoft Outlook Message 83 events 49 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
usc-excel.officeapps.live.com Document 24 events 6 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
gpteal.merck.com DOM 22 events 11 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.12
Swivel Rate
events in non-primary app / total
0.90
Zero-Edit Visit Rate
inspection-only clicks
1.21
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.90
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.90
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Swivel Rate
value 0.12
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Revisit Depth
value 1.21
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2354204 | Analyst 13 | 48.0 | 1.7m | 10 | 0.90 | 0.00 | 4 |
| 2353221 | Analyst 13 | 48.3 | 0.3m | 9 | 0.89 | 0.00 | 4 |
| 2329230 | Analyst 05 | 48.8 | 2.9m | 8 | 0.88 | 0.00 | 4 |
| 1677926 | Analyst 05 | 49.2 | 18.2m | 23 | 0.61 | 0.00 | 7 |
| 2341841 | Analyst 13 | 50.8 | 10.2m | 25 | 0.68 | 0.00 | 6 |
Users on This Task
15 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
06
15
07
04
14
10
16
11
01
03
02
09
12
05
13
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 06 | 2 | 2.0 | 90.0 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Save Changes
Analyst 06's Dominant Sequence
— 1 of 2 instances
1
Save Changes
2
Select List Item
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 4 | 4.0 | 87.1 | Zero-Edit Visits | 0.00 | 0.00 | 9 |
|
Reference SoP
1
Save Changes
Analyst 15's Dominant Sequence
— 3 of 4 instances
1
Save Changes
This analyst exhibits 2 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 7 | 7.0 | 86.8 | Zero-Edit Visits | 0.00 | 0.00 | 7 |
|
Reference SoP
1
Save Changes
Analyst 07's Dominant Sequence
— 3 of 7 instances
1
Save Changes
2
Preview English Narrative
This analyst exhibits 4 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 80 | 6.7 | 85.5 | Zero-Edit Visits | 0.00 | 0.00 | 110 |
|
Reference SoP
1
Save Changes
Analyst 04's Dominant Sequence
— 52 of 80 instances
1
Save Changes
This analyst exhibits 23 distinct sequences across their 80 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 12 | 6.0 | 83.7 | Zero-Edit Visits | 0.00 | 0.00 | 27 |
|
Reference SoP
1
Save Changes
Analyst 14's Dominant Sequence
— 4 of 12 instances
1
Save Changes
This analyst exhibits 9 distinct sequences across their 12 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Save Changes
Analyst 10's Dominant Sequence
— 1 of 1 instances
1
Save Changes
|
||||||||
| ▸ | Analyst 16 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 7 |
|
Reference SoP
1
Save Changes
Analyst 16's Dominant Sequence
— 1 of 1 instances
1
Save Changes
|
||||||||
| ▸ | Analyst 11 | 4 | 4.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 16 |
|
Reference SoP
1
Save Changes
Analyst 11's Dominant Sequence
— 3 of 4 instances
1
Save Changes
This analyst exhibits 2 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 39 | 2.8 | 81.0 | Zero-Edit Visits | 0.10 | 0.00 | 68 |
|
Reference SoP
1
Save Changes
Analyst 01's Dominant Sequence
— 12 of 39 instances
1
Save Changes
This analyst exhibits 28 distinct sequences across their 39 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 48 | 4.0 | 80.9 | Zero-Edit Visits | 0.09 | 0.00 | 77 |
|
Reference SoP
1
Save Changes
Analyst 03's Dominant Sequence
— 22 of 48 instances
1
Save Changes
This analyst exhibits 25 distinct sequences across their 48 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 45 | 3.8 | 78.1 | Zero-Edit Visits | 0.15 | 0.00 | 140 |
|
Reference SoP
1
Save Changes
Analyst 02's Dominant Sequence
— 8 of 45 instances
1
Save Changes
This analyst exhibits 36 distinct sequences across their 45 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 56 | 4.0 | 78.0 | Zero-Edit Visits | 0.17 | 0.00 | 207 |
|
Reference SoP
1
Save Changes
Analyst 09's Dominant Sequence
— 17 of 56 instances
1
Save Changes
This analyst exhibits 34 distinct sequences across their 56 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 63 | 7.0 | 77.1 | Zero-Edit Visits | 0.18 | 0.00 | 782 |
|
Reference SoP
1
Save Changes
Analyst 12's Dominant Sequence
— 18 of 63 instances
1
Save Changes
This analyst exhibits 41 distinct sequences across their 63 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 27 | 3.0 | 71.6 | Zero-Edit Visits | 0.30 | 0.00 | 113 |
|
Reference SoP
1
Save Changes
Analyst 05's Dominant Sequence
— 7 of 27 instances
1
Save Changes
This analyst exhibits 21 distinct sequences across their 27 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 15 | 2.1 | 69.9 | Zero-Edit Visits | 0.35 | 0.00 | 56 |
|
Reference SoP
1
Save Changes
Analyst 13's Dominant Sequence
— 6 of 15 instances
1
Save Changes
This analyst exhibits 10 distinct sequences across their 15 instances; the dominant one is shown.
|
||||||||
85.4
Avg Adherence Score (0-100)
Median 82.5 · Range 37.2–97.5
16.2 min
Avg Task Duration
0.3 min median
2.5
Avg Sub-Steps per Instance
75.4% have multiple sub-steps
14.3%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
14.3% of instances follow this exactly
Advance Case — Mined SoP
The most common sub-step sequence observed across 2,467 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Complete Action
1.1s
0.6 edits
Veeva Safety
66%
Variant Explorer
Reference
Advance Case Reference Path
353 instances · 10 analysts
This is the mined reference path — 353 instances across 10 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Advance Case.
Genuine Complexity
Complex-Case Execution Path
253 instances · 12 analysts
· Δ adherence +12.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Approval Complete
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 97.4 vs reference 85.4 (delta +12.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
165 instances · 5 analysts
· Δ adherence +11.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Data Entry Complete
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 97.4 vs reference 85.4 (delta +11.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
151 instances · 15 analysts
· Δ adherence -1.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Save Changes
reference at this step: Complete Action
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 84.0 vs reference 85.4 (delta -1.4). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
145 instances · 11 analysts
· Δ adherence -2.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Start Approval Workflow
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 82.7 vs reference 85.4 (delta -2.7). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
141 instances · 7 analysts
· Δ adherence -2.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Continue Workflow
reference at this step: Complete Action
2
Delete Item
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 82.6 vs reference 85.4 (delta -2.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
104 instances · 10 analysts
· Δ adherence -2.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Continue Workflow
reference at this step: Complete Action
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 83.4 vs reference 85.4 (delta -2.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
62 instances · 7 analysts
· Δ adherence +12.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Send to Case Approval
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 97.5 vs reference 85.4 (delta +12.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
61 instances · 3 analysts
· Δ adherence -1.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Open Cases Section
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 83.6 vs reference 85.4 (delta -1.8). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
33 instances · 6 analysts
· Δ adherence -1.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Navigate to Cases
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 83.6 vs reference 85.4 (delta -1.8). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
33 instances · 7 analysts
· Δ adherence +12.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Send for QC
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 97.5 vs reference 85.4 (delta +12.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
30 instances · 9 analysts
· Δ adherence -2.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Select Start Approval Workflow
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 82.5 vs reference 85.4 (delta -2.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
30 instances · 10 analysts
· Δ adherence +12.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Select Approval Complete
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 97.5 vs reference 85.4 (delta +12.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 5,407 | 86.5% | 98.6% |
| Microsoft Word | 160 | 2.6% | 0.0% |
| Acrobat | 146 | 2.3% | 0.0% |
| Microsoft Teams | 131 | 2.1% | 0.0% |
| collaboration.merck.com | 88 | 1.4% | 0.0% |
| Microsoft Excel | 80 | 1.3% | 0.0% |
| Microsoft Outlook | 76 | 1.2% | 0.0% |
| usc-excel.officeapps.live.com | 27 | 0.4% | 0.0% |
| gpteal.merck.com | 14 | 0.2% | 0.0% |
| Notepad | 12 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Veeva Safety | Microsoft Teams | 87 |
| Veeva Safety | Acrobat | 78 |
| Veeva Safety | Microsoft Word | 78 |
| Veeva Safety | Microsoft Outlook | 57 |
| Acrobat | Veeva Safety | 56 |
| Acrobat | Microsoft Word | 51 |
| Microsoft Teams | Veeva Safety | 49 |
| Veeva Safety | Microsoft Excel | 49 |
| Microsoft Word | Acrobat | 49 |
| Microsoft Word | Veeva Safety | 48 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 14,776 events 425 cases 232.6 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
1943×
pyzeClick
<a>
"Show more"
476×
pyzeClick
<span>
"Narrative"
470×
pyzeClick
<span>
"Continue"
447×
pyzeClick
<button>
"Save"
422×
pyzeClick
<li>
"Narrative"
395×
pyzeClick
<span>
"Adverse Events"
349×
pyzeClick
<li>
"Details"
325×
Phobos DOM 2,387 events 289 cases 72.5 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
359×
pyzeClick
<button>
"Advance"
225×
pyzeClick
<button>
"Save Comments"
183×
pyzeClick
<button>
"Send to process"
114×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
102×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
94×
pyzeClick
<a>
"Cases"
85×
pyzeClick
<button>
"Send to Medical Review"
68×
Microsoft Word Document 571 events 149 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 552979 (English) (26689105_0_2_348832)
1638×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
723×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
572×
file
V-Narrative 549550 (English) (26643764_0_1_347593)
526×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
382×
file
V-Narrative 533882 (English) (26397511_0_1_345751)
308×
file
V-Narrative 553023 (English) (26689066_0_2_351253)
257×
file
V-Narrative 542512 (English) (26522671_0_2_344761)
255×
Acrobat App 423 events 98 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
5,505 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 358 events 143 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
4,239 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Excel Document 202 events 81 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Case tracking list
696×
file
Daily tracking list (PV-SMART) copy
435×
file
1761620 (v2.1)
330×
file
Priorización
314×
file
2352822 (v0.1)
234×
file
E2B_MSDPRO_014172_AGPROD1_20251127163042
155×
file
I-0000353124
142×
file
I-0000350382
141×
Microsoft Outlook Message 194 events 108 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
collaboration.merck.com App 176 events 83 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Notepad App 50 events 27 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
ONENOTE Document 48 events 29 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.07
Swivel Rate
events in non-primary app / total
0.66
Zero-Edit Visit Rate
inspection-only clicks
1.09
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.66
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.66
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.09
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.07
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2309351 | Analyst 01 | 37.2 | 15.3m | 25 | 0.84 | 0.00 | 11 |
| 1677926 | Analyst 05 | 42.0 | 33.8m | 45 | 0.62 | 0.00 | 14 |
| 2345589 | Analyst 13 | 42.1 | 2.0m | 14 | 0.93 | 0.00 | 6 |
| 2235982 | Analyst 13 | 45.5 | 1.6m | 10 | 0.90 | 0.00 | 5 |
| 2341558 | Analyst 13 | 47.5 | 11.3m | 24 | 0.67 | 0.00 | 7 |
Users on This Task
15 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
06
15
14
07
04
11
01
03
02
09
05
12
10
16
13
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 06 | 6 | 3.0 | 88.8 | Zero-Edit Visits | 0.00 | 0.00 | 10 |
|
Reference SoP
1
Complete Action
Analyst 06's Dominant Sequence
— 1 of 6 instances
1
Complete Action
2
Start Approval Workflow
This analyst exhibits 6 distinct sequences across their 6 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 25 | 25.0 | 88.2 | Zero-Edit Visits | 0.00 | 0.00 | 36 |
|
Reference SoP
1
Complete Action
Analyst 15's Dominant Sequence
— 6 of 25 instances
1
Complete Action
2
Navigate to Cases
This analyst exhibits 15 distinct sequences across their 25 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 69 | 34.5 | 87.9 | Zero-Edit Visits | 0.00 | 0.00 | 85 |
|
Reference SoP
1
Complete Action
Analyst 14's Dominant Sequence
— 8 of 69 instances
1
Complete Action
2
Start Approval Workflow
This analyst exhibits 36 distinct sequences across their 69 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 17 | 17.0 | 87.5 | Zero-Edit Visits | 0.00 | 0.00 | 14 |
|
Reference SoP
1
Complete Action
Analyst 07's Dominant Sequence
— 3 of 17 instances
1
Save Changes
· reference: Complete Action
2
Preview English Narrative
This analyst exhibits 12 distinct sequences across their 17 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 401 | 28.6 | 87.4 | Zero-Edit Visits | 0.00 | 0.00 | 1,562 |
|
Reference SoP
1
Complete Action
Analyst 04's Dominant Sequence
— 56 of 401 instances
1
Complete Action
2
Open Cases Section
This analyst exhibits 95 distinct sequences across their 401 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 12 | 12.0 | 86.7 | Zero-Edit Visits | 0.00 | 0.00 | 22 |
|
Reference SoP
1
Complete Action
Analyst 11's Dominant Sequence
— 3 of 12 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 10 distinct sequences across their 12 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 351 | 23.4 | 86.4 | Zero-Edit Visits | 0.07 | 0.00 | 17,038 |
|
Reference SoP
1
Complete Action
Analyst 01's Dominant Sequence
— 77 of 351 instances
1
Complete Action
This analyst exhibits 112 distinct sequences across their 351 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 303 | 23.3 | 85.9 | Zero-Edit Visits | 0.05 | 0.00 | 4,387 |
|
Reference SoP
1
Complete Action
Analyst 03's Dominant Sequence
— 53 of 303 instances
1
Complete Action
This analyst exhibits 92 distinct sequences across their 303 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 353 | 29.4 | 85.5 | Zero-Edit Visits | 0.06 | 0.00 | 3,667 |
|
Reference SoP
1
Complete Action
Analyst 02's Dominant Sequence
— 77 of 353 instances
1
Complete Action
This analyst exhibits 128 distinct sequences across their 353 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 390 | 24.4 | 84.6 | Zero-Edit Visits | 0.08 | 0.00 | 9,142 |
|
Reference SoP
1
Complete Action
Analyst 09's Dominant Sequence
— 65 of 390 instances
1
Continue Workflow
· reference: Complete Action
2
Delete Item
This analyst exhibits 99 distinct sequences across their 390 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 191 | 19.1 | 83.9 | Zero-Edit Visits | 0.11 | 0.00 | 718 |
|
Reference SoP
1
Complete Action
Analyst 05's Dominant Sequence
— 27 of 191 instances
1
Complete Action
This analyst exhibits 91 distinct sequences across their 191 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 208 | 23.1 | 82.7 | Zero-Edit Visits | 0.11 | 0.00 | 1,868 |
|
Reference SoP
1
Complete Action
Analyst 12's Dominant Sequence
— 21 of 208 instances
1
Complete Action
This analyst exhibits 111 distinct sequences across their 208 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 2 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Complete Action
Analyst 10's Dominant Sequence
— 1 of 2 instances
1
Continue Workflow
· reference: Complete Action
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 16 | 2 | 2.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 7 |
|
Reference SoP
1
Complete Action
Analyst 16's Dominant Sequence
— 1 of 2 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 137 | 12.5 | 82.2 | Zero-Edit Visits | 0.17 | 0.00 | 1,351 |
|
Reference SoP
1
Complete Action
Analyst 13's Dominant Sequence
— 26 of 137 instances
1
Complete Action
This analyst exhibits 61 distinct sequences across their 137 instances; the dominant one is shown.
|
||||||||
88.0
Avg Adherence Score (0-100)
Median 82.5 · Range 42.1–97.5
19.1 min
Avg Task Duration
0.2 min median
2.1
Avg Sub-Steps per Instance
78.4% have multiple sub-steps
—
Reference Path Coverage
1,637 sub-task instances
Reference Baseline
Advance Case — Mined SoP
The most common sub-step sequence observed across 1,637 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Complete Action
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 3,193 | 91.0% | 98.3% |
| Microsoft Teams | 65 | 1.9% | 0.0% |
| Microsoft Outlook | 44 | 1.3% | 0.0% |
| Microsoft Excel | 43 | 1.2% | 0.0% |
| collaboration.merck.com | 42 | 1.2% | 0.0% |
| Acrobat | 29 | 0.8% | 0.0% |
| Microsoft Word | 22 | 0.6% | 0.0% |
| usc-excel.officeapps.live.com | 15 | 0.4% | 0.0% |
| gpteal.merck.com | 6 | 0.2% | 0.0% |
| apps.powerapps.com | 4 | 0.1% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Veeva Safety | Microsoft Teams | 46 |
| Veeva Safety | Microsoft Outlook | 33 |
| Veeva Safety | Microsoft Excel | 32 |
| Veeva Safety | Acrobat | 23 |
| Microsoft Teams | Veeva Safety | 22 |
| Microsoft Outlook | Veeva Safety | 19 |
| Veeva Safety | Microsoft Word | 17 |
| Veeva Safety | collaboration.merck.com | 17 |
| collaboration.merck.com | usc-excel.officeapps.live.com | 13 |
| Acrobat | Veeva Safety | 13 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 9,857 events 378 cases 132.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
1941×
pyzeClick
<a>
"Show more"
368×
pyzeClick
<span>
"Narrative"
311×
pyzeClick
<li>
"Narrative"
308×
pyzeClick
<span>
"Approval Complete"
306×
pyzeClick
<span>
"Start Approval Workflow"
252×
pyzeClick
<li>
"Details"
243×
pyzeClick
<span>
"Validation Results (Failures & Warnings)"
199×
Phobos DOM 1,934 events 255 cases 55.5 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
322×
pyzeClick
<button>
"Advance"
197×
pyzeClick
<button>
"Save Comments"
161×
pyzeClick
<button>
"Send to process"
88×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
87×
pyzeClick
<button>
"Send to Medical Review"
67×
pyzeClick
<a>
"Cases"
65×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
63×
Microsoft Word Document 329 events 129 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 528191 (English) (26300681_0_1_338349)
382×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
306×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
294×
file
V-Narrative 542512 (English) (26522671_0_2_344761)
255×
file
V-Narrative 531284 (English) (26355806_0_2_340584)
225×
file
V-Narrative 525091 (English) (26253532_0_1_338428)
201×
file
V-Narrative 531868 (English) (26362311_0_1_340534)
185×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
Microsoft Teams App 217 events 107 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,461 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Acrobat App 214 events 83 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,067 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Outlook Message 127 events 80 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
subj
RE: [Confidential] <Due on 24H> <Approved> <Correction multiple cases>Product Selection
32×
subj
<compensación> Case#2252065
31×
subj
<Due today><Void request><GPVCM POLAND> Case # 2326062
23×
subj
<GPVCM Announcement> Promotions to Safety Data Manager positions in Poland
18×
subj
<compensación> Case#2351734
17×
subj
Due today reports on 03-DEC-2025
14×
subj
Due today reports on 01-DEC-2025
13×
subj
RE: <compensación> Case#2309077
8×
Microsoft Excel Document 105 events 59 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
collaboration.merck.com App 98 events 52 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
ONENOTE Document 41 events 27 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Notepad App 33 events 18 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.05
Swivel Rate
events in non-primary app / total
0.53
Zero-Edit Visit Rate
inspection-only clicks
1.02
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.53
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.53
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.02
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.05
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2345589 | Analyst 13 | 42.1 | 2.0m | 14 | 0.93 | 0.00 | 6 |
| 2235982 | Analyst 13 | 45.5 | 1.6m | 10 | 0.90 | 0.00 | 5 |
| 2350053 | Analyst 02 | 50.1 | 4.5m | 9 | 0.89 | 0.30 | 1 |
| 2350576 | Analyst 13 | 51.0 | 0.4m | 10 | 0.80 | 0.00 | 4 |
| 2291397 | Analyst 13 | 56.0 | 5.4m | 5 | 0.80 | 0.00 | 2 |
Users on This Task
14 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
06
11
07
15
04
14
03
01
02
09
05
12
13
16
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 06 | 3 | 3.0 | 90.0 | Zero-Edit Visits | 0.00 | 0.00 | 8 |
|
Reference SoP
1
Complete Action
Analyst 06's Dominant Sequence
— 1 of 3 instances
1
Complete Action
2
Start Approval Workflow
This analyst exhibits 3 distinct sequences across their 3 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 7 | 7.0 | 89.6 | Zero-Edit Visits | 0.00 | 0.00 | 5 |
|
Reference SoP
1
Complete Action
Analyst 11's Dominant Sequence
— 1 of 7 instances
1
Complete Action
2
Select Send to Case Approval
This analyst exhibits 7 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 8 | 8.0 | 89.4 | Zero-Edit Visits | 0.00 | 0.00 | 5 |
|
Reference SoP
1
Complete Action
Analyst 07's Dominant Sequence
— 2 of 8 instances
1
Complete Action
2
Select Start Approval Workflow
This analyst exhibits 6 distinct sequences across their 8 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 19 | 19.0 | 89.1 | Zero-Edit Visits | 0.00 | 0.00 | 15 |
|
Reference SoP
1
Complete Action
Analyst 15's Dominant Sequence
— 6 of 19 instances
1
Complete Action
2
Navigate to Cases
This analyst exhibits 10 distinct sequences across their 19 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 255 | 18.2 | 89.0 | Zero-Edit Visits | 0.00 | 0.00 | 1,221 |
|
Reference SoP
1
Complete Action
Analyst 04's Dominant Sequence
— 56 of 255 instances
1
Complete Action
2
Open Cases Section
This analyst exhibits 45 distinct sequences across their 255 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 52 | 26.0 | 89.0 | Zero-Edit Visits | 0.00 | 0.00 | 36 |
|
Reference SoP
1
Complete Action
Analyst 14's Dominant Sequence
— 8 of 52 instances
1
Complete Action
2
Start Approval Workflow
This analyst exhibits 22 distinct sequences across their 52 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 204 | 15.7 | 88.4 | Zero-Edit Visits | 0.04 | 0.00 | 3,049 |
|
Reference SoP
1
Complete Action
Analyst 03's Dominant Sequence
— 53 of 204 instances
1
Complete Action
This analyst exhibits 48 distinct sequences across their 204 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 260 | 17.3 | 88.2 | Zero-Edit Visits | 0.07 | 0.00 | 16,808 |
|
Reference SoP
1
Complete Action
Analyst 01's Dominant Sequence
— 77 of 260 instances
1
Complete Action
This analyst exhibits 58 distinct sequences across their 260 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 240 | 20.0 | 88.1 | Zero-Edit Visits | 0.05 | 0.00 | 202 |
|
Reference SoP
1
Complete Action
Analyst 02's Dominant Sequence
— 77 of 240 instances
1
Complete Action
This analyst exhibits 63 distinct sequences across their 240 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 228 | 14.2 | 88.1 | Zero-Edit Visits | 0.05 | 0.00 | 7,311 |
|
Reference SoP
1
Complete Action
Analyst 09's Dominant Sequence
— 42 of 228 instances
1
Complete Action
2
Mark Approval Complete
This analyst exhibits 36 distinct sequences across their 228 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 144 | 14.4 | 87.2 | Zero-Edit Visits | 0.06 | 0.00 | 439 |
|
Reference SoP
1
Complete Action
Analyst 05's Dominant Sequence
— 27 of 144 instances
1
Complete Action
This analyst exhibits 54 distinct sequences across their 144 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 111 | 15.9 | 86.9 | Zero-Edit Visits | 0.07 | 0.00 | 1,001 |
|
Reference SoP
1
Complete Action
Analyst 12's Dominant Sequence
— 21 of 111 instances
1
Complete Action
This analyst exhibits 45 distinct sequences across their 111 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 105 | 10.5 | 85.1 | Zero-Edit Visits | 0.13 | 0.00 | 1,100 |
|
Reference SoP
1
Complete Action
Analyst 13's Dominant Sequence
— 26 of 105 instances
1
Complete Action
This analyst exhibits 40 distinct sequences across their 105 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 16 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Complete Action
Analyst 16's Dominant Sequence
— 1 of 1 instances
1
Complete Action
|
||||||||
81.0
Avg Adherence Score (0-100)
Median 82.5 · Range 37.2–97.5
16.1 min
Avg Task Duration
0.3 min median
3.1
Avg Sub-Steps per Instance
75.6% have multiple sub-steps
—
Reference Path Coverage
426 sub-task instances
Reference Baseline
Advance Case — Mined SoP
The most common sub-step sequence observed across 426 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Continue Workflow
2
Notepad
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 1,104 | 84.8% | 99.0% |
| Acrobat | 58 | 4.5% | 0.0% |
| Microsoft Word | 57 | 4.4% | 0.0% |
| Microsoft Teams | 25 | 1.9% | 0.0% |
| Microsoft Outlook | 10 | 0.8% | 0.0% |
| Microsoft Excel | 9 | 0.7% | 0.0% |
| collaboration.merck.com | 8 | 0.6% | 0.0% |
| gpteal.merck.com | 6 | 0.5% | 0.0% |
| TextInputHost | 3 | 0.2% | 0.0% |
| YoudaoDict | 2 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Acrobat | Microsoft Word | 29 |
| Veeva Safety | Acrobat | 26 |
| Microsoft Word | Acrobat | 26 |
| Acrobat | Veeva Safety | 20 |
| Veeva Safety | Microsoft Word | 19 |
| Veeva Safety | Microsoft Teams | 17 |
| Microsoft Word | Veeva Safety | 16 |
| Microsoft Teams | Veeva Safety | 9 |
| Veeva Safety | Microsoft Outlook | 7 |
| Veeva Safety | Microsoft Excel | 6 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 4,654 events 167 cases 77.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Continue"
447×
pyzeClick
<span>
"Complete"
327×
pyzeClick
<li>
"Delete"
300×
pyzeClick
<button>
"Create"
151×
pyzeClick
<a>
"Show more"
118×
pyzeClick
<li>
"Narrative"
112×
pyzeClick
<span>
"Narrative"
103×
pyzeClick
<li>
"Details"
92×
Phobos DOM 305 events 57 cases 11.0 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
57×
pyzeClick
<button>
"Advance"
33×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
24×
pyzeClick
<button>
"Save Comments"
18×
pyzeClick
<a>
"Cases"
17×
pyzeClick
<button>
">> Send to Quality Check"
13×
pyzeClick
<a>
"Cases Coordinator"
11×
pyzeClick
<button>
"Send to process"
10×
Microsoft Word Document 189 events 48 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 552979 (English) (26689105_0_2_348832)
557×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
515×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
429×
file
V-Narrative 553023 (English) (26689066_0_2_351253)
251×
file
V-Narrative 533882 (English) (26397511_0_1_345751)
230×
file
V-Narrative 549428 (English) (26642399_0_2_349187)
170×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
file
V-Narrative 538518 (English) (26459147_0_1_344628)
169×
Acrobat App 154 events 38 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,365 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 109 events 50 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,019 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Excel Document 54 events 24 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
1761620 (v2.1)
330×
file
2352822 (v0.1)
216×
file
Case tracking list
168×
file
MERCK2516915_ARGPRO_20251125054618
97×
file
Daily tracking list (PV-SMART) copy
87×
file
Priorización
68×
file
ICHICSR-EVHUMAN_10020432161 (1)
30×
file
GDMS800-039-001 GPVCM Finding Definitions (1)
20×
Microsoft Outlook Message 51 events 34 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
collaboration.merck.com App 36 events 21 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Notepad App 19 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
gpteal.merck.com DOM 18 events 10 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.07
Swivel Rate
events in non-primary app / total
0.92
Zero-Edit Visit Rate
inspection-only clicks
1.2
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.92
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.92
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.20
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.07
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2309351 | Analyst 01 | 37.2 | 15.3m | 25 | 0.84 | 0.00 | 11 |
| 2341558 | Analyst 13 | 47.5 | 11.3m | 24 | 0.67 | 0.00 | 7 |
| 2252065 | Analyst 05 | 49.4 | 5.4m | 16 | 0.69 | 0.00 | 6 |
| 2350507 | Analyst 09 | 50.0 | 13.7m | 12 | 0.75 | 0.00 | 5 |
| 2314923 | Analyst 05 | 54.2 | 20.6m | 26 | 0.46 | 0.00 | 7 |
Users on This Task
14 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
14
04
06
15
07
10
11
01
02
03
09
12
05
13
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 14 | 5 | 5.0 | 86.2 | Zero-Edit Visits | 0.00 | 0.00 | 10 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 14's Dominant Sequence
— 1 of 5 instances
1
Continue Workflow
2
Press Enter
· reference: Notepad
3
Continue Link
4
Select List Item
5
View Study
6
Select Change Type
7
View External Product
This analyst exhibits 5 distinct sequences across their 5 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 66 | 6.0 | 83.7 | Zero-Edit Visits | 0.00 | 0.00 | 189 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 04's Dominant Sequence
— 28 of 66 instances
1
Continue Workflow
This analyst exhibits 27 distinct sequences across their 66 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 06 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 06's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
|
||||||||
| ▸ | Analyst 15 | 2 | 2.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 15's Dominant Sequence
— 1 of 2 instances
1
Continue Workflow
2
View Drug History
· reference: Notepad
3
View Adverse Events
4
View Products
5
View Case Assessments
6
View Case Assessment Results
7
Select Generate Narrative
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 2 | 2.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 07's Dominant Sequence
— 1 of 2 instances
1
Continue Workflow
2
Show More Details
· reference: Notepad
3
Select Reference Numbers
4
Delete Item
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 10's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
|
||||||||
| ▸ | Analyst 11 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 11's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
2
View Details Menu
· reference: Notepad
|
||||||||
| ▸ | Analyst 01 | 52 | 3.7 | 81.5 | Zero-Edit Visits | 0.08 | 0.00 | 129 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 01's Dominant Sequence
— 16 of 52 instances
1
Continue Workflow
This analyst exhibits 25 distinct sequences across their 52 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 68 | 6.8 | 81.3 | Zero-Edit Visits | 0.07 | 0.00 | 3,317 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 02's Dominant Sequence
— 32 of 68 instances
1
Continue Workflow
2
Delete Item
· reference: Notepad
This analyst exhibits 28 distinct sequences across their 68 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 51 | 3.9 | 80.8 | Zero-Edit Visits | 0.06 | 0.00 | 1,261 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 03's Dominant Sequence
— 22 of 51 instances
1
Continue Workflow
This analyst exhibits 19 distinct sequences across their 51 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 106 | 8.8 | 80.7 | Zero-Edit Visits | 0.07 | 0.00 | 1,607 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 09's Dominant Sequence
— 65 of 106 instances
1
Continue Workflow
2
Delete Item
· reference: Notepad
This analyst exhibits 28 distinct sequences across their 106 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 34 | 6.8 | 79.6 | Zero-Edit Visits | 0.12 | 0.00 | 85 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 12's Dominant Sequence
— 8 of 34 instances
1
Continue Workflow
2
Delete Item
· reference: Notepad
This analyst exhibits 25 distinct sequences across their 34 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 20 | 2.9 | 76.4 | Zero-Edit Visits | 0.20 | 0.00 | 89 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 05's Dominant Sequence
— 4 of 20 instances
1
Continue Workflow
This analyst exhibits 15 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 17 | 1.9 | 75.5 | Zero-Edit Visits | 0.22 | 0.00 | 181 |
|
Reference SoP
1
Continue Workflow
2
Notepad
Analyst 13's Dominant Sequence
— 8 of 17 instances
1
Continue Workflow
This analyst exhibits 10 distinct sequences across their 17 instances; the dominant one is shown.
|
||||||||
79.7
Avg Adherence Score (0-100)
Median 82.5 · Range 42.0–97.5
4.5 min
Avg Task Duration
0.7 min median
3.6
Avg Sub-Steps per Instance
62.6% have multiple sub-steps
—
Reference Path Coverage
404 sub-task instances
Reference Baseline
Advance Case — Mined SoP
The most common sub-step sequence observed across 404 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Save Changes
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 1,110 | 77.0% | 100.0% |
| Microsoft Word | 81 | 5.6% | 0.0% |
| Acrobat | 59 | 4.1% | 0.0% |
| Microsoft Teams | 41 | 2.8% | 0.0% |
| collaboration.merck.com | 38 | 2.6% | 0.0% |
| Microsoft Excel | 28 | 1.9% | 0.0% |
| Microsoft Outlook | 22 | 1.5% | 0.0% |
| usc-excel.officeapps.live.com | 12 | 0.8% | 0.0% |
| Notepad | 8 | 0.6% | 0.0% |
| evocpr01.apps.americas.cf.merck.com | 6 | 0.4% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Veeva Safety | Microsoft Word | 40 |
| Veeva Safety | Acrobat | 32 |
| Microsoft Word | Veeva Safety | 26 |
| Acrobat | Veeva Safety | 23 |
| Veeva Safety | Microsoft Teams | 23 |
| Acrobat | Microsoft Word | 21 |
| Microsoft Word | Acrobat | 20 |
| Microsoft Teams | Veeva Safety | 19 |
| Veeva Safety | collaboration.merck.com | 19 |
| Veeva Safety | Microsoft Outlook | 15 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 6,667 events 254 cases 125.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
449×
pyzeClick
<button>
"Save"
422×
pyzeClick
<span>
"Narrative"
217×
pyzeClick
<li>
"Narrative"
183×
pyzeClick
<a>
"Show more"
183×
pyzeClick
<span>
"Adverse Events"
179×
pyzeClick
<li>
"Details"
149×
pyzeClick
<button>
"Create"
133×
Phobos DOM 754 events 111 cases 26.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
105×
pyzeClick
<button>
"Advance"
71×
pyzeClick
<button>
"Save Comments"
59×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
52×
pyzeClick
<button>
"Send to process"
40×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
24×
pyzeClick
<a>
"Cases"
22×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
19×
Microsoft Word Document 281 events 81 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 552979 (English) (26689105_0_2_348832)
1076×
file
V-Narrative 549550 (English) (26643764_0_1_347593)
526×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
515×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
377×
file
V-Narrative 533909 (English) (26397745_0_2_338171)
243×
file
V-Narrative 545329 (English) (26567747_0_2_344796)
215×
file
V-Narrative 540358 (English) (26494608_0_2_342069)
195×
file
Narrative 526208 (English)
190×
Acrobat App 207 events 57 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
3,088 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 163 events 68 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,132 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Excel Document 110 events 44 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Case tracking list
382×
file
2352822 (v0.1)
234×
file
Daily tracking list (PV-SMART) copy
210×
file
I-0000353124
142×
file
I-0000350382
141×
file
Priorización
61×
file
MERCK2524493_ARGPRO_20251202045619
46×
file
ICHICSR-EVHUMAN_10020432161 (1)
37×
collaboration.merck.com App 100 events 48 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Microsoft Outlook Message 83 events 49 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
usc-excel.officeapps.live.com Document 24 events 6 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Notepad App 22 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.12
Swivel Rate
events in non-primary app / total
0.90
Zero-Edit Visit Rate
inspection-only clicks
1.25
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.90
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.90
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Swivel Rate
value 0.12
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Revisit Depth
value 1.25
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 1677926 | Analyst 05 | 42.0 | 33.8m | 45 | 0.62 | 0.00 | 14 |
| 2354204 | Analyst 13 | 48.0 | 1.7m | 10 | 0.90 | 0.00 | 4 |
| 2353221 | Analyst 13 | 48.3 | 0.3m | 9 | 0.89 | 0.00 | 4 |
| 2329230 | Analyst 05 | 48.8 | 2.9m | 8 | 0.88 | 0.00 | 4 |
| 2341841 | Analyst 13 | 50.8 | 10.2m | 25 | 0.68 | 0.00 | 6 |
Users on This Task
15 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
06
07
15
04
14
11
10
16
01
03
02
09
12
05
13
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 06 | 2 | 2.0 | 90.0 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Save Changes
Analyst 06's Dominant Sequence
— 1 of 2 instances
1
Save Changes
2
Select List Item
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 7 | 7.0 | 86.8 | Zero-Edit Visits | 0.00 | 0.00 | 7 |
|
Reference SoP
1
Save Changes
Analyst 07's Dominant Sequence
— 3 of 7 instances
1
Save Changes
2
Preview English Narrative
This analyst exhibits 4 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 4 | 4.0 | 86.5 | Zero-Edit Visits | 0.00 | 0.00 | 17 |
|
Reference SoP
1
Save Changes
Analyst 15's Dominant Sequence
— 2 of 4 instances
1
Save Changes
This analyst exhibits 3 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 80 | 6.7 | 85.5 | Zero-Edit Visits | 0.00 | 0.00 | 153 |
|
Reference SoP
1
Save Changes
Analyst 04's Dominant Sequence
— 51 of 80 instances
1
Save Changes
This analyst exhibits 24 distinct sequences across their 80 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 12 | 6.0 | 84.0 | Zero-Edit Visits | 0.00 | 0.00 | 39 |
|
Reference SoP
1
Save Changes
Analyst 14's Dominant Sequence
— 4 of 12 instances
1
Save Changes
This analyst exhibits 9 distinct sequences across their 12 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 4 | 4.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 16 |
|
Reference SoP
1
Save Changes
Analyst 11's Dominant Sequence
— 3 of 4 instances
1
Save Changes
This analyst exhibits 2 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Save Changes
Analyst 10's Dominant Sequence
— 1 of 1 instances
1
Save Changes
|
||||||||
| ▸ | Analyst 16 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 7 |
|
Reference SoP
1
Save Changes
Analyst 16's Dominant Sequence
— 1 of 1 instances
1
Save Changes
|
||||||||
| ▸ | Analyst 01 | 39 | 2.8 | 80.9 | Zero-Edit Visits | 0.10 | 0.00 | 100 |
|
Reference SoP
1
Save Changes
Analyst 01's Dominant Sequence
— 11 of 39 instances
1
Save Changes
This analyst exhibits 29 distinct sequences across their 39 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 48 | 4.0 | 80.9 | Zero-Edit Visits | 0.09 | 0.00 | 77 |
|
Reference SoP
1
Save Changes
Analyst 03's Dominant Sequence
— 22 of 48 instances
1
Save Changes
This analyst exhibits 25 distinct sequences across their 48 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 45 | 3.8 | 78.1 | Zero-Edit Visits | 0.16 | 0.00 | 148 |
|
Reference SoP
1
Save Changes
Analyst 02's Dominant Sequence
— 8 of 45 instances
1
Save Changes
This analyst exhibits 36 distinct sequences across their 45 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 56 | 4.0 | 78.0 | Zero-Edit Visits | 0.17 | 0.00 | 224 |
|
Reference SoP
1
Save Changes
Analyst 09's Dominant Sequence
— 16 of 56 instances
1
Save Changes
This analyst exhibits 35 distinct sequences across their 56 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 63 | 7.0 | 77.1 | Zero-Edit Visits | 0.18 | 0.00 | 782 |
|
Reference SoP
1
Save Changes
Analyst 12's Dominant Sequence
— 18 of 63 instances
1
Save Changes
This analyst exhibits 41 distinct sequences across their 63 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 27 | 3.0 | 71.2 | Zero-Edit Visits | 0.30 | 0.00 | 189 |
|
Reference SoP
1
Save Changes
Analyst 05's Dominant Sequence
— 6 of 27 instances
1
Save Changes
This analyst exhibits 22 distinct sequences across their 27 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 15 | 2.1 | 69.4 | Zero-Edit Visits | 0.37 | 0.00 | 70 |
|
Reference SoP
1
Save Changes
Analyst 13's Dominant Sequence
— 5 of 15 instances
1
Save Changes
This analyst exhibits 11 distinct sequences across their 15 instances; the dominant one is shown.
|
||||||||
81.4
Avg Adherence Score (0-100)
Median 82.5 · Range 40.2–97.5
31.4 min
Avg Task Duration
1.0 min median
5.1
Avg Sub-Steps per Instance
75.6% have multiple sub-steps
7.7%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
7.7% of instances follow this exactly
Advance Case — Mined SoP
The most common sub-step sequence observed across 1,198 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Complete Action
1.5s
0.8 edits
Veeva Safety
53%
Variant Explorer
Reference
Advance Case Reference Path
92 instances · 11 analysts
This is the mined reference path — 92 instances across 11 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Advance Case.
Genuine Complexity
Complex-Case Execution Path
125 instances · 14 analysts
· Δ adherence +2.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Save Changes
reference at this step: Complete Action
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 84.2 vs reference 81.4 (delta +2.8). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
75 instances · 10 analysts
· Δ adherence +1.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Continue Workflow
reference at this step: Complete Action
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 83.3 vs reference 81.4 (delta +1.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Amendment Spiral
Post-Amendment Rework
49 instances · 6 analysts
· Δ adherence +9.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Approval Complete
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
Amendment Spiral — multiple analysts (6) revisit the same activities on the same cases (49), consistent with rework triggered by case version bumps (v2.0 → v3.0). Suggests either upstream intake issues driving amendments, or downstream rework after review feedback.
Amendment Spiral
Post-Amendment Rework
27 instances · 4 analysts
· Δ adherence +8.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Data Entry Complete
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
Amendment Spiral — multiple analysts (4) revisit the same activities on the same cases (27), consistent with rework triggered by case version bumps (v2.0 → v3.0). Suggests either upstream intake issues driving amendments, or downstream rework after review feedback.
Backfill Loop
Mid-Flow Data Backfill
18 instances · 2 analysts
· Δ adherence +2.2
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Send to Case Approval
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
4
Start Approval Workflow
extra step (reference ends at step 1)
5
Complete Action
extra step (reference ends at step 1)
6
Mark Approval Complete
extra step (reference ends at step 1)
7
Complete Action
extra step (reference ends at step 1)
Backfill Loop — instances revisit the same activity 4.0× on average, consistent with mid-flow detours to fetch or enter data that wasn't ready at the right step. Ordering issue, not an app issue — 2 analysts across 18 cases. Suggests upstream data readiness or UX sequencing is at fault.
Backfill Loop
Mid-Flow Data Backfill
15 instances · 1 analysts
· Δ adherence +1.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Data Entry Complete
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
4
Start Approval Workflow
extra step (reference ends at step 1)
5
Complete Action
extra step (reference ends at step 1)
6
Mark Approval Complete
extra step (reference ends at step 1)
7
Complete Action
extra step (reference ends at step 1)
8
Open Cases Section
extra step (reference ends at step 1)
Backfill Loop — instances revisit the same activity 4.0× on average, consistent with mid-flow detours to fetch or enter data that wasn't ready at the right step. Ordering issue, not an app issue — 1 analysts across 15 cases. Suggests upstream data readiness or UX sequencing is at fault.
Backfill Loop
Mid-Flow Data Backfill
14 instances · 4 analysts
· Δ adherence +2.2
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Data Entry Complete
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
4
Start Approval Workflow
extra step (reference ends at step 1)
5
Complete Action
extra step (reference ends at step 1)
6
Mark Approval Complete
extra step (reference ends at step 1)
7
Complete Action
extra step (reference ends at step 1)
Backfill Loop — instances revisit the same activity 4.0× on average, consistent with mid-flow detours to fetch or enter data that wasn't ready at the right step. Ordering issue, not an app issue — 4 analysts across 14 cases. Suggests upstream data readiness or UX sequencing is at fault.
Amendment Spiral
Post-Amendment Rework
14 instances · 6 analysts
· Δ adherence +9.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Send for QC
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
Amendment Spiral — multiple analysts (6) revisit the same activities on the same cases (14), consistent with rework triggered by case version bumps (v2.0 → v3.0). Suggests either upstream intake issues driving amendments, or downstream rework after review feedback.
Genuine Complexity
Complex-Case Execution Path
13 instances · 2 analysts
· Δ adherence +4.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Open Cases Section
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 85.4 vs reference 81.4 (delta +4.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Amendment Spiral
Post-Amendment Rework
12 instances · 3 analysts
· Δ adherence +6.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Data Entry Complete
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
4
Start Approval Workflow
extra step (reference ends at step 1)
Amendment Spiral — multiple analysts (3) revisit the same activities on the same cases (12), consistent with rework triggered by case version bumps (v2.0 → v3.0). Suggests either upstream intake issues driving amendments, or downstream rework after review feedback.
Genuine Complexity
Complex-Case Execution Path
10 instances · 1 analysts
· Δ adherence +6.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Send for QC
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
4
Open Cases Section
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (4 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 87.5 vs reference 81.4 (delta +6.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
8 instances · 5 analysts
· Δ adherence +8.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Cancel Action
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 90.0 vs reference 81.4 (delta +8.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 5,258 | 86.8% | 98.8% |
| Microsoft Word | 149 | 2.5% | 0.0% |
| Acrobat | 130 | 2.1% | 0.0% |
| Microsoft Teams | 126 | 2.1% | 0.0% |
| collaboration.merck.com | 86 | 1.4% | 0.0% |
| Microsoft Excel | 79 | 1.3% | 0.0% |
| Microsoft Outlook | 74 | 1.2% | 0.0% |
| usc-excel.officeapps.live.com | 25 | 0.4% | 0.0% |
| gpteal.merck.com | 14 | 0.2% | 0.0% |
| Notepad | 12 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Veeva Safety | Microsoft Teams | 84 |
| Veeva Safety | Acrobat | 74 |
| Veeva Safety | Microsoft Word | 69 |
| Microsoft Teams | Veeva Safety | 58 |
| Veeva Safety | Microsoft Outlook | 56 |
| Acrobat | Veeva Safety | 52 |
| Veeva Safety | Microsoft Excel | 50 |
| Acrobat | Microsoft Word | 48 |
| Microsoft Word | Veeva Safety | 48 |
| Microsoft Word | Acrobat | 41 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 14,776 events 425 cases 232.6 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
1943×
pyzeClick
<a>
"Show more"
476×
pyzeClick
<span>
"Narrative"
470×
pyzeClick
<span>
"Continue"
447×
pyzeClick
<button>
"Save"
422×
pyzeClick
<li>
"Narrative"
395×
pyzeClick
<span>
"Adverse Events"
349×
pyzeClick
<li>
"Details"
325×
Phobos DOM 2,387 events 289 cases 72.5 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
359×
pyzeClick
<button>
"Advance"
225×
pyzeClick
<button>
"Save Comments"
183×
pyzeClick
<button>
"Send to process"
114×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
102×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
94×
pyzeClick
<a>
"Cases"
85×
pyzeClick
<button>
"Send to Medical Review"
68×
Microsoft Word Document 571 events 149 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 552979 (English) (26689105_0_2_348832)
1638×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
723×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
572×
file
V-Narrative 549550 (English) (26643764_0_1_347593)
526×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
382×
file
V-Narrative 533882 (English) (26397511_0_1_345751)
308×
file
V-Narrative 553023 (English) (26689066_0_2_351253)
257×
file
V-Narrative 542512 (English) (26522671_0_2_344761)
255×
Acrobat App 423 events 98 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
5,505 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 358 events 143 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
4,239 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Excel Document 202 events 81 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Case tracking list
696×
file
Daily tracking list (PV-SMART) copy
435×
file
1761620 (v2.1)
330×
file
Priorización
314×
file
2352822 (v0.1)
234×
file
E2B_MSDPRO_014172_AGPROD1_20251127163042
155×
file
I-0000353124
142×
file
I-0000350382
141×
Microsoft Outlook Message 194 events 108 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
collaboration.merck.com App 176 events 83 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Notepad App 50 events 27 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
ONENOTE Document 48 events 29 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.09
Swivel Rate
events in non-primary app / total
0.71
Zero-Edit Visit Rate
inspection-only clicks
2.11
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.72
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.72
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 2.11
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.09
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2309351 | Analyst 01 | 40.2 | 17.0m | 28 | 0.75 | 0.00 | 11 |
| 2353221 | Analyst 13 | 48.3 | 0.3m | 9 | 0.89 | 0.00 | 4 |
| 2329230 | Analyst 05 | 48.8 | 2.9m | 8 | 0.88 | 0.00 | 4 |
| 2341558 | Analyst 13 | 50.8 | 1041.4m | 27 | 0.59 | 0.00 | 7 |
| 2341841 | Analyst 13 | 50.8 | 10.2m | 25 | 0.68 | 0.00 | 6 |
Users on This Task
15 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
06
15
04
01
11
10
16
02
14
03
07
12
09
13
05
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 06 | 4 | 2.0 | 89.4 | Zero-Edit Visits | 0.00 | 0.00 | 10 |
|
Reference SoP
1
Complete Action
Analyst 06's Dominant Sequence
— 1 of 4 instances
1
Complete Action
This analyst exhibits 4 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 11 | 11.0 | 84.5 | Zero-Edit Visits | 0.00 | 0.00 | 28 |
|
Reference SoP
1
Complete Action
Analyst 15's Dominant Sequence
— 3 of 11 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 9 distinct sequences across their 11 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 203 | 14.5 | 83.9 | Zero-Edit Visits | 0.00 | 0.00 | 938 |
|
Reference SoP
1
Complete Action
Analyst 04's Dominant Sequence
— 40 of 203 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 88 distinct sequences across their 203 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 202 | 13.5 | 83.9 | Zero-Edit Visits | 0.09 | 0.00 | 16,983 |
|
Reference SoP
1
Complete Action
Analyst 01's Dominant Sequence
— 17 of 202 instances
1
Complete Action
2
Mark Approval Complete
3
Complete Action
This analyst exhibits 113 distinct sequences across their 202 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 7 | 7.0 | 83.8 | Zero-Edit Visits | 0.00 | 0.00 | 22 |
|
Reference SoP
1
Complete Action
Analyst 11's Dominant Sequence
— 3 of 7 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 5 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 2 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Complete Action
Analyst 10's Dominant Sequence
— 1 of 2 instances
1
Continue Workflow
· reference: Complete Action
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 16 | 2 | 2.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 7 |
|
Reference SoP
1
Complete Action
Analyst 16's Dominant Sequence
— 1 of 2 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 178 | 14.8 | 82.3 | Zero-Edit Visits | 0.08 | 0.00 | 3,626 |
|
Reference SoP
1
Complete Action
Analyst 02's Dominant Sequence
— 22 of 178 instances
1
Complete Action
2
Mark Approval Complete
3
Complete Action
This analyst exhibits 111 distinct sequences across their 178 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 26 | 13.0 | 82.1 | Zero-Edit Visits | 0.00 | 0.00 | 77 |
|
Reference SoP
1
Complete Action
Analyst 14's Dominant Sequence
— 4 of 26 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 22 distinct sequences across their 26 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 150 | 11.5 | 81.7 | Zero-Edit Visits | 0.08 | 0.00 | 4,397 |
|
Reference SoP
1
Complete Action
Analyst 03's Dominant Sequence
— 16 of 150 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 86 distinct sequences across their 150 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 6 | 6.0 | 81.4 | Zero-Edit Visits | 0.00 | 0.00 | 15 |
|
Reference SoP
1
Complete Action
Analyst 07's Dominant Sequence
— 2 of 6 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 5 distinct sequences across their 6 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 109 | 12.1 | 78.5 | Zero-Edit Visits | 0.16 | 0.00 | 1,869 |
|
Reference SoP
1
Complete Action
Analyst 12's Dominant Sequence
— 14 of 109 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 84 distinct sequences across their 109 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 148 | 9.2 | 78.3 | Zero-Edit Visits | 0.14 | 0.00 | 7,628 |
|
Reference SoP
1
Complete Action
Analyst 09's Dominant Sequence
— 17 of 148 instances
1
Complete Action
2
Send to Case Approval
3
Complete Action
4
Start Approval Workflow
5
Complete Action
6
Mark Approval Complete
7
Complete Action
This analyst exhibits 85 distinct sequences across their 148 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 69 | 6.3 | 77.8 | Zero-Edit Visits | 0.20 | 0.00 | 1,338 |
|
Reference SoP
1
Complete Action
Analyst 13's Dominant Sequence
— 8 of 69 instances
1
Continue Workflow
· reference: Complete Action
This analyst exhibits 46 distinct sequences across their 69 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 81 | 8.1 | 77.4 | Zero-Edit Visits | 0.15 | 0.00 | 632 |
|
Reference SoP
1
Complete Action
Analyst 05's Dominant Sequence
— 10 of 81 instances
1
Complete Action
This analyst exhibits 64 distinct sequences across their 81 instances; the dominant one is shown.
|
||||||||
83.6
Avg Adherence Score (0-100)
Median 83.3 · Range 52.6–97.5
45.7 min
Avg Task Duration
0.9 min median
5.2
Avg Sub-Steps per Instance
85.5% have multiple sub-steps
—
Reference Path Coverage
635 sub-task instances
Reference Baseline
Advance Case — Mined SoP
The most common sub-step sequence observed across 635 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Complete Action
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 3,019 | 91.1% | 98.1% |
| Microsoft Teams | 59 | 1.8% | 0.0% |
| Microsoft Excel | 42 | 1.3% | 0.0% |
| Microsoft Outlook | 41 | 1.2% | 0.0% |
| collaboration.merck.com | 36 | 1.1% | 0.0% |
| Acrobat | 29 | 0.9% | 0.0% |
| Microsoft Word | 23 | 0.7% | 0.0% |
| usc-excel.officeapps.live.com | 15 | 0.5% | 0.0% |
| gpteal.merck.com | 5 | 0.2% | 0.0% |
| app.powerbi.com | 5 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Veeva Safety | Microsoft Teams | 43 |
| Veeva Safety | Microsoft Excel | 32 |
| Veeva Safety | Microsoft Outlook | 31 |
| Microsoft Teams | Veeva Safety | 29 |
| Veeva Safety | Acrobat | 21 |
| Microsoft Outlook | Veeva Safety | 21 |
| Veeva Safety | Microsoft Word | 19 |
| Acrobat | Veeva Safety | 17 |
| Veeva Safety | collaboration.merck.com | 15 |
| collaboration.merck.com | usc-excel.officeapps.live.com | 13 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 9,857 events 378 cases 132.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
1941×
pyzeClick
<a>
"Show more"
368×
pyzeClick
<span>
"Narrative"
311×
pyzeClick
<li>
"Narrative"
308×
pyzeClick
<span>
"Approval Complete"
306×
pyzeClick
<span>
"Start Approval Workflow"
252×
pyzeClick
<li>
"Details"
243×
pyzeClick
<span>
"Data Entry Complete"
199×
Phobos DOM 1,934 events 255 cases 55.5 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
322×
pyzeClick
<button>
"Advance"
197×
pyzeClick
<button>
"Save Comments"
161×
pyzeClick
<button>
"Send to process"
88×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
87×
pyzeClick
<button>
"Send to Medical Review"
67×
pyzeClick
<a>
"Cases"
65×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
63×
Microsoft Word Document 329 events 129 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 528191 (English) (26300681_0_1_338349)
382×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
306×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
294×
file
V-Narrative 542512 (English) (26522671_0_2_344761)
255×
file
V-Narrative 531284 (English) (26355806_0_2_340584)
225×
file
V-Narrative 525091 (English) (26253532_0_1_338428)
201×
file
V-Narrative 531868 (English) (26362311_0_1_340534)
185×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
Microsoft Teams App 217 events 107 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,461 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Acrobat App 214 events 83 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,067 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Outlook Message 127 events 80 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
subj
RE: [Confidential] <Due on 24H> <Approved> <Correction multiple cases>Product Selection
32×
subj
<compensación> Case#2252065
31×
subj
<Due today><Void request><GPVCM POLAND> Case # 2326062
23×
subj
<GPVCM Announcement> Promotions to Safety Data Manager positions in Poland
18×
subj
<compensación> Case#2351734
17×
subj
Due today reports on 03-DEC-2025
14×
subj
Due today reports on 01-DEC-2025
13×
subj
RE: <compensación> Case#2309077
8×
Microsoft Excel Document 105 events 59 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
collaboration.merck.com App 98 events 52 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
ONENOTE Document 41 events 27 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Notepad App 33 events 18 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.07
Swivel Rate
events in non-primary app / total
0.56
Zero-Edit Visit Rate
inspection-only clicks
2.44
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.56
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.56
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 2.44
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.07
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2345589 | Analyst 13 | 52.6 | 6.2m | 25 | 0.68 | 0.00 | 6 |
| 2312686 | Analyst 12 | 56.5 | 33.7m | 10 | 0.70 | 0.00 | 3 |
| 2353949 | Analyst 01 | 57.5 | 4.2m | 4 | 0.75 | 0.00 | 2 |
| 2235982 | Analyst 13 | 57.9 | 2.4m | 21 | 0.52 | 0.00 | 6 |
| 2351742 | Analyst 12 | 60.0 | 0.7m | 6 | 0.67 | 0.00 | 2 |
Users on This Task
14 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
06
01
11
02
03
04
15
16
14
12
07
13
09
05
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 06 | 2 | 2.0 | 92.5 | Zero-Edit Visits | 0.00 | 0.00 | 8 |
|
Reference SoP
1
Complete Action
Analyst 06's Dominant Sequence
— 1 of 2 instances
1
Complete Action
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 134 | 8.9 | 85.7 | Zero-Edit Visits | 0.08 | 0.00 | 16,800 |
|
Reference SoP
1
Complete Action
Analyst 01's Dominant Sequence
— 17 of 134 instances
1
Complete Action
2
Mark Approval Complete
3
Complete Action
This analyst exhibits 62 distinct sequences across their 134 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 3 | 3.0 | 85.4 | Zero-Edit Visits | 0.00 | 0.00 | 6 |
|
Reference SoP
1
Complete Action
Analyst 11's Dominant Sequence
— 1 of 3 instances
1
Complete Action
2
Select Send to Case Approval
3
Complete Action
4
Select Start Approval Workflow
5
Complete Action
6
Select Approval Complete
7
Complete Action
8
Navigate to Inbox
This analyst exhibits 3 distinct sequences across their 3 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 112 | 9.3 | 84.5 | Zero-Edit Visits | 0.06 | 0.00 | 215 |
|
Reference SoP
1
Complete Action
Analyst 02's Dominant Sequence
— 23 of 112 instances
1
Complete Action
2
Mark Approval Complete
3
Complete Action
This analyst exhibits 59 distinct sequences across their 112 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 77 | 5.9 | 83.6 | Zero-Edit Visits | 0.06 | 0.00 | 3,055 |
|
Reference SoP
1
Complete Action
Analyst 03's Dominant Sequence
— 16 of 77 instances
1
Complete Action
This analyst exhibits 46 distinct sequences across their 77 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 91 | 6.5 | 83.5 | Zero-Edit Visits | 0.00 | 0.00 | 159 |
|
Reference SoP
1
Complete Action
Analyst 04's Dominant Sequence
— 18 of 91 instances
1
Complete Action
This analyst exhibits 39 distinct sequences across their 91 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 6 | 6.0 | 83.4 | Zero-Edit Visits | 0.00 | 0.00 | 16 |
|
Reference SoP
1
Complete Action
Analyst 15's Dominant Sequence
— 1 of 6 instances
1
Complete Action
2
Send to Case Approval
3
Complete Action
4
Start Approval Workflow
5
Complete Link
6
Mark Approval Complete
7
Complete Action
8
Navigate to Cases
This analyst exhibits 6 distinct sequences across their 6 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 16 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Complete Action
Analyst 16's Dominant Sequence
— 1 of 1 instances
1
Complete Action
|
||||||||
| ▸ | Analyst 14 | 14 | 7.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 38 |
|
Reference SoP
1
Complete Action
Analyst 14's Dominant Sequence
— 2 of 14 instances
1
Complete Action
This analyst exhibits 13 distinct sequences across their 14 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 40 | 5.7 | 82.4 | Zero-Edit Visits | 0.10 | 0.00 | 902 |
|
Reference SoP
1
Complete Action
Analyst 12's Dominant Sequence
— 4 of 40 instances
1
Complete Action
This analyst exhibits 35 distinct sequences across their 40 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 1 | 1.0 | 81.7 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Complete Action
Analyst 07's Dominant Sequence
— 1 of 1 instances
1
Complete Action
2
Select Send to Case Approval
3
Complete Action
4
Select Start Approval Workflow
5
Complete Action
6
Mark Approval Complete
7
View Required Fields Dialog
8
Complete Action
9
Navigate to Cases
|
||||||||
| ▸ | Analyst 13 | 39 | 3.9 | 81.7 | Zero-Edit Visits | 0.14 | 0.00 | 71 |
|
Reference SoP
1
Complete Action
Analyst 13's Dominant Sequence
— 5 of 39 instances
1
Complete Action
This analyst exhibits 28 distinct sequences across their 39 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 70 | 4.4 | 81.1 | Zero-Edit Visits | 0.10 | 0.00 | 7,315 |
|
Reference SoP
1
Complete Action
Analyst 09's Dominant Sequence
— 17 of 70 instances
1
Complete Action
2
Send to Case Approval
3
Complete Action
4
Start Approval Workflow
5
Complete Action
6
Mark Approval Complete
7
Complete Action
This analyst exhibits 34 distinct sequences across their 70 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 45 | 4.5 | 81.1 | Zero-Edit Visits | 0.06 | 0.00 | 436 |
|
Reference SoP
1
Complete Action
Analyst 05's Dominant Sequence
— 10 of 45 instances
1
Complete Action
This analyst exhibits 34 distinct sequences across their 45 instances; the dominant one is shown.
|
||||||||
78.7
Avg Adherence Score (0-100)
Median 82.5 · Range 40.2–97.5
32.0 min
Avg Task Duration
1.3 min median
5.8
Avg Sub-Steps per Instance
65.0% have multiple sub-steps
—
Reference Path Coverage
214 sub-task instances
Reference Baseline
Advance Case — Mined SoP
The most common sub-step sequence observed across 214 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 1,044 | 84.2% | 100.0% |
| Microsoft Word | 53 | 4.3% | 0.0% |
| Acrobat | 51 | 4.1% | 0.0% |
| Microsoft Teams | 27 | 2.2% | 0.0% |
| collaboration.merck.com | 13 | 1.0% | 0.0% |
| Microsoft Outlook | 10 | 0.8% | 0.0% |
| Microsoft Excel | 7 | 0.6% | 0.0% |
| gpteal.merck.com | 6 | 0.5% | 0.0% |
| Google Search | 3 | 0.2% | 0.0% |
| Notepad | 3 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Acrobat | Microsoft Word | 27 |
| Microsoft Word | Acrobat | 26 |
| Veeva Safety | Acrobat | 22 |
| Veeva Safety | Microsoft Teams | 20 |
| Veeva Safety | Microsoft Word | 17 |
| Acrobat | Veeva Safety | 16 |
| Microsoft Word | Veeva Safety | 12 |
| Microsoft Teams | Veeva Safety | 10 |
| Veeva Safety | Microsoft Outlook | 7 |
| Veeva Safety | collaboration.merck.com | 5 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 4,654 events 167 cases 77.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Continue"
447×
pyzeClick
<span>
"Complete"
327×
pyzeClick
<li>
"Delete"
300×
pyzeClick
<button>
"Create"
151×
pyzeClick
<a>
"Show more"
118×
pyzeClick
<li>
"Narrative"
112×
pyzeClick
<span>
"Narrative"
103×
pyzeClick
<li>
"Details"
92×
Phobos DOM 305 events 57 cases 11.0 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
57×
pyzeClick
<button>
"Advance"
33×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
24×
pyzeClick
<button>
"Save Comments"
18×
pyzeClick
<a>
"Cases"
17×
pyzeClick
<button>
">> Send to Quality Check"
13×
pyzeClick
<a>
"Cases Coordinator"
11×
pyzeClick
<button>
"Send to process"
10×
Microsoft Word Document 189 events 48 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 552979 (English) (26689105_0_2_348832)
557×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
515×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
429×
file
V-Narrative 553023 (English) (26689066_0_2_351253)
251×
file
V-Narrative 533882 (English) (26397511_0_1_345751)
230×
file
V-Narrative 549428 (English) (26642399_0_2_349187)
170×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
file
V-Narrative 538518 (English) (26459147_0_1_344628)
169×
Acrobat App 154 events 38 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,365 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 109 events 50 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,019 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Excel Document 54 events 24 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
1761620 (v2.1)
330×
file
2352822 (v0.1)
216×
file
Case tracking list
168×
file
MERCK2516915_ARGPRO_20251125054618
97×
file
Daily tracking list (PV-SMART) copy
87×
file
Priorización
68×
file
ICHICSR-EVHUMAN_10020432161 (1)
30×
file
GDMS800-039-001 GPVCM Finding Definitions (1)
20×
Microsoft Outlook Message 51 events 34 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
collaboration.merck.com App 36 events 21 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Notepad App 19 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
gpteal.merck.com DOM 18 events 10 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.11
Swivel Rate
events in non-primary app / total
0.89
Zero-Edit Visit Rate
inspection-only clicks
2.12
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.89
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.89
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 2.12
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.11
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2309351 | Analyst 01 | 40.2 | 17.0m | 28 | 0.75 | 0.00 | 11 |
| 2341558 | Analyst 13 | 50.8 | 1041.4m | 27 | 0.59 | 0.00 | 7 |
| 2252065 | Analyst 05 | 53.6 | 6.7m | 21 | 0.57 | 0.00 | 6 |
| 2350507 | Analyst 09 | 53.8 | 14.7m | 16 | 0.69 | 0.00 | 5 |
| 2347303 | Analyst 02 | 55.9 | 5.7m | 11 | 0.64 | 0.00 | 4 |
Users on This Task
13 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
04
10
15
06
02
07
01
03
05
14
12
13
09
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 04 | 42 | 4.2 | 82.8 | Zero-Edit Visits | 0.00 | 0.00 | 623 |
|
Reference SoP
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 04's Dominant Sequence
— 18 of 42 instances
1
Continue Workflow
· reference: Delete Item
This analyst exhibits 23 distinct sequences across their 42 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 10's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
· reference: Delete Item
|
||||||||
| ▸ | Analyst 15 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 15's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
· reference: Delete Item
2
View Details
· reference: Continue Workflow
|
||||||||
| ▸ | Analyst 06 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 06's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
· reference: Delete Item
|
||||||||
| ▸ | Analyst 02 | 29 | 2.9 | 80.2 | Zero-Edit Visits | 0.10 | 0.00 | 3,278 |
|
Reference SoP
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 02's Dominant Sequence
— 8 of 29 instances
1
Continue Workflow
· reference: Delete Item
This analyst exhibits 22 distinct sequences across their 29 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 1 | 1.0 | 80.0 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 07's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
· reference: Delete Item
2
Show More Details
· reference: Continue Workflow
3
Select Reference Numbers
· reference: Google Search
4
Delete Item
· reference: pubmed.ncbi.nlm.nih.gov
5
Continue Workflow
· reference: Google Search
6
View Validation Results
· reference: Navigate to Cases
7
Select Products
· reference: Create New Item
|
||||||||
| ▸ | Analyst 01 | 32 | 2.3 | 79.7 | Zero-Edit Visits | 0.10 | 0.00 | 107 |
|
Reference SoP
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 01's Dominant Sequence
— 9 of 32 instances
1
Continue Workflow
· reference: Delete Item
This analyst exhibits 22 distinct sequences across their 32 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 32 | 2.7 | 78.7 | Zero-Edit Visits | 0.10 | 0.00 | 1,263 |
|
Reference SoP
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 03's Dominant Sequence
— 15 of 32 instances
1
Continue Workflow
· reference: Delete Item
This analyst exhibits 17 distinct sequences across their 32 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 16 | 2.3 | 75.9 | Zero-Edit Visits | 0.21 | 0.00 | 77 |
|
Reference SoP
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 05's Dominant Sequence
— 2 of 16 instances
1
Continue Workflow
· reference: Delete Item
This analyst exhibits 15 distinct sequences across their 16 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 1 | 1.0 | 75.9 | Zero-Edit Visits | 0.00 | 0.00 | 10 |
|
Reference SoP
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 14's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
· reference: Delete Item
2
Press Enter
· reference: Continue Workflow
3
Continue Link
· reference: Google Search
4
Select List Item
· reference: pubmed.ncbi.nlm.nih.gov
5
View Study
· reference: Google Search
6
Select Change Type
· reference: Navigate to Cases
7
View External Product
· reference: Create New Item
8
Continue Workflow
· reference: Select External Product
9
Select Change Type
· reference: Continue
10
View External Product
· reference: Select Products
11
Continue Workflow
· reference: Delete Item
12
Select Change Type
· reference: Continue Workflow
13
View External Product
· reference: Notepad
14
Continue Link
15
Continue Workflow
16
View Products
17
Select Change Type
18
View External Product
19
Continue Workflow
20
View Case Assessments
21
Select Not Applicable
22
View Case Assessment Results
|
||||||||
| ▸ | Analyst 12 | 14 | 2.8 | 75.8 | Zero-Edit Visits | 0.19 | 0.00 | 180 |
|
Reference SoP
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 12's Dominant Sequence
— 3 of 14 instances
1
Continue Workflow
· reference: Delete Item
This analyst exhibits 12 distinct sequences across their 14 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 16 | 1.8 | 75.6 | Zero-Edit Visits | 0.21 | 0.00 | 1,209 |
|
Reference SoP
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 13's Dominant Sequence
— 8 of 16 instances
1
Continue Workflow
· reference: Delete Item
This analyst exhibits 9 distinct sequences across their 16 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 28 | 2.3 | 74.2 | Zero-Edit Visits | 0.17 | 0.00 | 94 |
|
Reference SoP
1
Delete Item
2
Continue Workflow
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 09's Dominant Sequence
— 10 of 28 instances
1
Continue Workflow
· reference: Delete Item
This analyst exhibits 19 distinct sequences across their 28 instances; the dominant one is shown.
|
||||||||
79.1
Avg Adherence Score (0-100)
Median 82.5 · Range 48.3–97.5
4.9 min
Avg Task Duration
0.9 min median
4.3
Avg Sub-Steps per Instance
64.2% have multiple sub-steps
—
Reference Path Coverage
349 sub-task instances
Reference Baseline
Advance Case — Mined SoP
The most common sub-step sequence observed across 349 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Save Changes
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 1,195 | 79.5% | 100.0% |
| Microsoft Word | 73 | 4.9% | 0.0% |
| Acrobat | 50 | 3.3% | 0.0% |
| Microsoft Teams | 40 | 2.7% | 0.0% |
| collaboration.merck.com | 37 | 2.5% | 0.0% |
| Microsoft Excel | 30 | 2.0% | 0.0% |
| Microsoft Outlook | 23 | 1.5% | 0.0% |
| usc-excel.officeapps.live.com | 10 | 0.7% | 0.0% |
| Notepad | 5 | 0.3% | 0.0% |
| ONENOTE | 5 | 0.3% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Veeva Safety | Microsoft Word | 35 |
| Veeva Safety | Acrobat | 28 |
| Microsoft Word | Veeva Safety | 23 |
| Veeva Safety | Microsoft Teams | 23 |
| Acrobat | Veeva Safety | 21 |
| Veeva Safety | collaboration.merck.com | 20 |
| Microsoft Teams | Veeva Safety | 19 |
| Acrobat | Microsoft Word | 18 |
| Veeva Safety | Microsoft Outlook | 16 |
| Microsoft Word | Acrobat | 15 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 6,667 events 254 cases 125.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
449×
pyzeClick
<button>
"Save"
422×
pyzeClick
<span>
"Narrative"
217×
pyzeClick
<li>
"Narrative"
183×
pyzeClick
<a>
"Show more"
183×
pyzeClick
<span>
"Adverse Events"
179×
pyzeClick
<li>
"Details"
149×
pyzeClick
<button>
"Create"
133×
Phobos DOM 754 events 111 cases 26.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
105×
pyzeClick
<button>
"Advance"
71×
pyzeClick
<button>
"Save Comments"
59×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
52×
pyzeClick
<button>
"Send to process"
40×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
24×
pyzeClick
<a>
"Cases"
22×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
19×
Microsoft Word Document 281 events 81 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 552979 (English) (26689105_0_2_348832)
1076×
file
V-Narrative 549550 (English) (26643764_0_1_347593)
526×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
515×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
377×
file
V-Narrative 533909 (English) (26397745_0_2_338171)
243×
file
V-Narrative 545329 (English) (26567747_0_2_344796)
215×
file
V-Narrative 540358 (English) (26494608_0_2_342069)
195×
file
Narrative 526208 (English)
190×
Acrobat App 207 events 57 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
3,088 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 163 events 68 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,132 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Excel Document 110 events 44 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Case tracking list
382×
file
2352822 (v0.1)
234×
file
Daily tracking list (PV-SMART) copy
210×
file
I-0000353124
142×
file
I-0000350382
141×
file
Priorización
61×
file
MERCK2524493_ARGPRO_20251202045619
46×
file
ICHICSR-EVHUMAN_10020432161 (1)
37×
collaboration.merck.com App 100 events 48 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Microsoft Outlook Message 83 events 49 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
usc-excel.officeapps.live.com Document 24 events 6 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Notepad App 22 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.12
Swivel Rate
events in non-primary app / total
0.90
Zero-Edit Visit Rate
inspection-only clicks
1.5
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.90
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.90
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Swivel Rate
value 0.13
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Revisit Depth
value 1.50
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2353221 | Analyst 13 | 48.3 | 0.3m | 9 | 0.89 | 0.00 | 4 |
| 2329230 | Analyst 05 | 48.8 | 2.9m | 8 | 0.88 | 0.00 | 4 |
| 2341841 | Analyst 13 | 50.8 | 10.2m | 25 | 0.68 | 0.00 | 6 |
| 2329808 | Analyst 13 | 51.0 | 6.9m | 10 | 0.80 | 0.00 | 4 |
| 2344413 | Analyst 03 | 51.8 | 3.6m | 7 | 0.86 | 0.00 | 3 |
Users on This Task
15 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
06
15
04
16
11
10
14
07
01
03
02
09
12
05
13
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 06 | 1 | 1.0 | 90.0 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Save Changes
Analyst 06's Dominant Sequence
— 1 of 1 instances
1
Save Changes
2
Select List Item
3
Save Changes
|
||||||||
| ▸ | Analyst 15 | 4 | 4.0 | 86.7 | Zero-Edit Visits | 0.00 | 0.00 | 11 |
|
Reference SoP
1
Save Changes
Analyst 15's Dominant Sequence
— 3 of 4 instances
1
Save Changes
This analyst exhibits 2 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 70 | 5.8 | 85.2 | Zero-Edit Visits | 0.00 | 0.00 | 155 |
|
Reference SoP
1
Save Changes
Analyst 04's Dominant Sequence
— 40 of 70 instances
1
Save Changes
This analyst exhibits 25 distinct sequences across their 70 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 16 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 7 |
|
Reference SoP
1
Save Changes
Analyst 16's Dominant Sequence
— 1 of 1 instances
1
Save Changes
|
||||||||
| ▸ | Analyst 11 | 4 | 4.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 16 |
|
Reference SoP
1
Save Changes
Analyst 11's Dominant Sequence
— 3 of 4 instances
1
Save Changes
This analyst exhibits 2 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Save Changes
Analyst 10's Dominant Sequence
— 1 of 1 instances
1
Save Changes
|
||||||||
| ▸ | Analyst 14 | 11 | 5.5 | 82.2 | Zero-Edit Visits | 0.00 | 0.00 | 29 |
|
Reference SoP
1
Save Changes
Analyst 14's Dominant Sequence
— 4 of 11 instances
1
Save Changes
This analyst exhibits 8 distinct sequences across their 11 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 4 | 4.0 | 81.6 | Zero-Edit Visits | 0.00 | 0.00 | 12 |
|
Reference SoP
1
Save Changes
Analyst 07's Dominant Sequence
— 2 of 4 instances
1
Save Changes
This analyst exhibits 3 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 36 | 2.6 | 80.8 | Zero-Edit Visits | 0.10 | 0.00 | 76 |
|
Reference SoP
1
Save Changes
Analyst 01's Dominant Sequence
— 8 of 36 instances
1
Save Changes
This analyst exhibits 29 distinct sequences across their 36 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 41 | 3.4 | 80.4 | Zero-Edit Visits | 0.09 | 0.00 | 79 |
|
Reference SoP
1
Save Changes
Analyst 03's Dominant Sequence
— 16 of 41 instances
1
Save Changes
This analyst exhibits 24 distinct sequences across their 41 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 37 | 3.4 | 77.4 | Zero-Edit Visits | 0.15 | 0.00 | 134 |
|
Reference SoP
1
Save Changes
Analyst 02's Dominant Sequence
— 6 of 37 instances
1
Save Changes
This analyst exhibits 30 distinct sequences across their 37 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 50 | 3.6 | 76.8 | Zero-Edit Visits | 0.19 | 0.00 | 219 |
|
Reference SoP
1
Save Changes
Analyst 09's Dominant Sequence
— 15 of 50 instances
1
Save Changes
This analyst exhibits 32 distinct sequences across their 50 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 55 | 6.1 | 76.5 | Zero-Edit Visits | 0.20 | 0.00 | 786 |
|
Reference SoP
1
Save Changes
Analyst 12's Dominant Sequence
— 14 of 55 instances
1
Save Changes
This analyst exhibits 37 distinct sequences across their 55 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 20 | 2.2 | 70.4 | Zero-Edit Visits | 0.29 | 0.00 | 119 |
|
Reference SoP
1
Save Changes
Analyst 05's Dominant Sequence
— 6 of 20 instances
1
Save Changes
This analyst exhibits 15 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 14 | 2.0 | 69.5 | Zero-Edit Visits | 0.34 | 0.00 | 58 |
|
Reference SoP
1
Save Changes
Analyst 13's Dominant Sequence
— 6 of 14 instances
1
Save Changes
This analyst exhibits 9 distinct sequences across their 14 instances; the dominant one is shown.
|
||||||||
81.3
Avg Adherence Score (0-100)
Median 82.5 · Range 40.2–97.5
33.8 min
Avg Task Duration
1.0 min median
5.3
Avg Sub-Steps per Instance
77.3% have multiple sub-steps
7.1%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
7.1% of instances follow this exactly
Advance Case — Mined SoP
The most common sub-step sequence observed across 1,184 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Complete Action
1.6s
0.8 edits
Veeva Safety
53%
Variant Explorer
Reference
Advance Case Reference Path
84 instances · 9 analysts
This is the mined reference path — 84 instances across 9 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Advance Case.
Genuine Complexity
Complex-Case Execution Path
118 instances · 13 analysts
· Δ adherence +3.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Save Changes
reference at this step: Complete Action
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 84.3 vs reference 81.3 (delta +3.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
67 instances · 9 analysts
· Δ adherence +1.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Continue Workflow
reference at this step: Complete Action
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 83.2 vs reference 81.3 (delta +1.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Amendment Spiral
Post-Amendment Rework
48 instances · 6 analysts
· Δ adherence +9.2
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Approval Complete
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
Amendment Spiral — multiple analysts (6) revisit the same activities on the same cases (48), consistent with rework triggered by case version bumps (v2.0 → v3.0). Suggests either upstream intake issues driving amendments, or downstream rework after review feedback.
Amendment Spiral
Post-Amendment Rework
27 instances · 4 analysts
· Δ adherence +8.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Data Entry Complete
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
Amendment Spiral — multiple analysts (4) revisit the same activities on the same cases (27), consistent with rework triggered by case version bumps (v2.0 → v3.0). Suggests either upstream intake issues driving amendments, or downstream rework after review feedback.
Backfill Loop
Mid-Flow Data Backfill
18 instances · 2 analysts
· Δ adherence +2.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Send to Case Approval
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
4
Start Approval Workflow
extra step (reference ends at step 1)
5
Complete Action
extra step (reference ends at step 1)
6
Mark Approval Complete
extra step (reference ends at step 1)
7
Complete Action
extra step (reference ends at step 1)
Backfill Loop — instances revisit the same activity 4.0× on average, consistent with mid-flow detours to fetch or enter data that wasn't ready at the right step. Ordering issue, not an app issue — 2 analysts across 18 cases. Suggests upstream data readiness or UX sequencing is at fault.
Genuine Complexity
Complex-Case Execution Path
15 instances · 2 analysts
· Δ adherence +3.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Open Cases Section
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 85.0 vs reference 81.3 (delta +3.7). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Backfill Loop
Mid-Flow Data Backfill
14 instances · 4 analysts
· Δ adherence +2.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Data Entry Complete
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
4
Start Approval Workflow
extra step (reference ends at step 1)
5
Complete Action
extra step (reference ends at step 1)
6
Mark Approval Complete
extra step (reference ends at step 1)
7
Complete Action
extra step (reference ends at step 1)
Backfill Loop — instances revisit the same activity 4.0× on average, consistent with mid-flow detours to fetch or enter data that wasn't ready at the right step. Ordering issue, not an app issue — 4 analysts across 14 cases. Suggests upstream data readiness or UX sequencing is at fault.
Amendment Spiral
Post-Amendment Rework
14 instances · 6 analysts
· Δ adherence +9.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Send for QC
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
Amendment Spiral — multiple analysts (6) revisit the same activities on the same cases (14), consistent with rework triggered by case version bumps (v2.0 → v3.0). Suggests either upstream intake issues driving amendments, or downstream rework after review feedback.
Backfill Loop
Mid-Flow Data Backfill
14 instances · 1 analysts
· Δ adherence +1.2
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Data Entry Complete
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
4
Start Approval Workflow
extra step (reference ends at step 1)
5
Complete Action
extra step (reference ends at step 1)
6
Mark Approval Complete
extra step (reference ends at step 1)
7
Complete Action
extra step (reference ends at step 1)
8
Open Cases Section
extra step (reference ends at step 1)
Backfill Loop — instances revisit the same activity 4.0× on average, consistent with mid-flow detours to fetch or enter data that wasn't ready at the right step. Ordering issue, not an app issue — 1 analysts across 14 cases. Suggests upstream data readiness or UX sequencing is at fault.
Amendment Spiral
Post-Amendment Rework
12 instances · 3 analysts
· Δ adherence +6.2
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Mark Data Entry Complete
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
4
Start Approval Workflow
extra step (reference ends at step 1)
Amendment Spiral — multiple analysts (3) revisit the same activities on the same cases (12), consistent with rework triggered by case version bumps (v2.0 → v3.0). Suggests either upstream intake issues driving amendments, or downstream rework after review feedback.
Genuine Complexity
Complex-Case Execution Path
9 instances · 1 analysts
· Δ adherence +6.2
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Send for QC
extra step (reference ends at step 1)
3
Complete Action
extra step (reference ends at step 1)
4
Open Cases Section
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (4 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 87.5 vs reference 81.3 (delta +6.2). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
8 instances · 5 analysts
· Δ adherence +8.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Complete Action
2
Cancel Action
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 90.0 vs reference 81.3 (delta +8.7). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 5,407 | 86.5% | 98.6% |
| Microsoft Word | 160 | 2.6% | 0.0% |
| Acrobat | 146 | 2.3% | 0.0% |
| Microsoft Teams | 131 | 2.1% | 0.0% |
| collaboration.merck.com | 88 | 1.4% | 0.0% |
| Microsoft Excel | 80 | 1.3% | 0.0% |
| Microsoft Outlook | 76 | 1.2% | 0.0% |
| usc-excel.officeapps.live.com | 27 | 0.4% | 0.0% |
| gpteal.merck.com | 14 | 0.2% | 0.0% |
| Notepad | 12 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Veeva Safety | Microsoft Teams | 86 |
| Veeva Safety | Acrobat | 81 |
| Veeva Safety | Microsoft Word | 75 |
| Acrobat | Veeva Safety | 62 |
| Microsoft Teams | Veeva Safety | 60 |
| Veeva Safety | Microsoft Outlook | 58 |
| Microsoft Word | Veeva Safety | 55 |
| Acrobat | Microsoft Word | 53 |
| Veeva Safety | Microsoft Excel | 50 |
| Microsoft Word | Acrobat | 47 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 14,776 events 425 cases 232.6 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
1943×
pyzeClick
<a>
"Show more"
476×
pyzeClick
<span>
"Narrative"
470×
pyzeClick
<span>
"Continue"
447×
pyzeClick
<button>
"Save"
422×
pyzeClick
<li>
"Narrative"
395×
pyzeClick
<span>
"Adverse Events"
349×
pyzeClick
<li>
"Details"
325×
Phobos DOM 2,387 events 289 cases 72.5 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
359×
pyzeClick
<button>
"Advance"
225×
pyzeClick
<button>
"Save Comments"
183×
pyzeClick
<button>
"Send to process"
114×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
102×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
94×
pyzeClick
<a>
"Cases"
85×
pyzeClick
<button>
"Send to Medical Review"
68×
Microsoft Word Document 571 events 149 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 552979 (English) (26689105_0_2_348832)
1638×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
723×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
572×
file
V-Narrative 549550 (English) (26643764_0_1_347593)
526×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
382×
file
V-Narrative 533882 (English) (26397511_0_1_345751)
308×
file
V-Narrative 553023 (English) (26689066_0_2_351253)
257×
file
V-Narrative 542512 (English) (26522671_0_2_344761)
255×
Acrobat App 423 events 98 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
5,505 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 358 events 143 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
4,239 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Excel Document 202 events 81 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Case tracking list
696×
file
Daily tracking list (PV-SMART) copy
435×
file
1761620 (v2.1)
330×
file
Priorización
314×
file
2352822 (v0.1)
234×
file
E2B_MSDPRO_014172_AGPROD1_20251127163042
155×
file
I-0000353124
142×
file
I-0000350382
141×
Microsoft Outlook Message 194 events 108 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
collaboration.merck.com App 176 events 83 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Notepad App 50 events 27 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
ONENOTE Document 48 events 29 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.09
Swivel Rate
events in non-primary app / total
0.71
Zero-Edit Visit Rate
inspection-only clicks
2.14
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.71
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.71
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 2.14
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.09
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2309351 | Analyst 01 | 40.2 | 17.0m | 28 | 0.75 | 0.00 | 11 |
| 1677926 | Analyst 05 | 43.4 | 34.9m | 48 | 0.58 | 0.00 | 14 |
| 2353221 | Analyst 13 | 48.3 | 0.3m | 9 | 0.89 | 0.00 | 4 |
| 2329230 | Analyst 05 | 48.8 | 2.9m | 8 | 0.88 | 0.00 | 4 |
| 2341841 | Analyst 13 | 50.8 | 10.2m | 25 | 0.68 | 0.00 | 6 |
Users on This Task
15 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
06
15
04
01
11
10
16
14
02
03
07
12
09
13
05
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 06 | 3 | 1.5 | 88.3 | Zero-Edit Visits | 0.00 | 0.00 | 11 |
|
Reference SoP
1
Complete Action
Analyst 06's Dominant Sequence
— 1 of 3 instances
1
Complete Action
2
Start Approval Workflow
3
Continue Workflow
This analyst exhibits 3 distinct sequences across their 3 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 10 | 10.0 | 84.5 | Zero-Edit Visits | 0.00 | 0.00 | 40 |
|
Reference SoP
1
Complete Action
Analyst 15's Dominant Sequence
— 2 of 10 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 9 distinct sequences across their 10 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 201 | 14.4 | 83.9 | Zero-Edit Visits | 0.00 | 0.00 | 1,595 |
|
Reference SoP
1
Complete Action
Analyst 04's Dominant Sequence
— 39 of 201 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 94 distinct sequences across their 201 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 201 | 13.4 | 83.9 | Zero-Edit Visits | 0.09 | 0.00 | 17,046 |
|
Reference SoP
1
Complete Action
Analyst 01's Dominant Sequence
— 17 of 201 instances
1
Complete Action
2
Mark Approval Complete
3
Complete Action
This analyst exhibits 113 distinct sequences across their 201 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 7 | 7.0 | 83.8 | Zero-Edit Visits | 0.00 | 0.00 | 22 |
|
Reference SoP
1
Complete Action
Analyst 11's Dominant Sequence
— 3 of 7 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 5 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 2 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Complete Action
Analyst 10's Dominant Sequence
— 1 of 2 instances
1
Continue Workflow
· reference: Complete Action
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 16 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 8 |
|
Reference SoP
1
Complete Action
Analyst 16's Dominant Sequence
— 1 of 1 instances
1
Save Changes
· reference: Complete Action
2
Complete Action
|
||||||||
| ▸ | Analyst 14 | 26 | 13.0 | 82.3 | Zero-Edit Visits | 0.00 | 0.00 | 90 |
|
Reference SoP
1
Complete Action
Analyst 14's Dominant Sequence
— 4 of 26 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 22 distinct sequences across their 26 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 177 | 14.8 | 82.2 | Zero-Edit Visits | 0.09 | 0.00 | 3,676 |
|
Reference SoP
1
Complete Action
Analyst 02's Dominant Sequence
— 22 of 177 instances
1
Complete Action
2
Mark Approval Complete
3
Complete Action
This analyst exhibits 111 distinct sequences across their 177 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 148 | 11.4 | 81.6 | Zero-Edit Visits | 0.08 | 0.00 | 4,413 |
|
Reference SoP
1
Complete Action
Analyst 03's Dominant Sequence
— 16 of 148 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 87 distinct sequences across their 148 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 6 | 6.0 | 81.4 | Zero-Edit Visits | 0.00 | 0.00 | 15 |
|
Reference SoP
1
Complete Action
Analyst 07's Dominant Sequence
— 2 of 6 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 5 distinct sequences across their 6 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 109 | 12.1 | 78.5 | Zero-Edit Visits | 0.16 | 0.00 | 1,876 |
|
Reference SoP
1
Complete Action
Analyst 12's Dominant Sequence
— 14 of 109 instances
1
Save Changes
· reference: Complete Action
This analyst exhibits 85 distinct sequences across their 109 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 147 | 9.2 | 78.2 | Zero-Edit Visits | 0.14 | 0.00 | 9,159 |
|
Reference SoP
1
Complete Action
Analyst 09's Dominant Sequence
— 17 of 147 instances
1
Complete Action
2
Send to Case Approval
3
Complete Action
4
Start Approval Workflow
5
Complete Action
6
Mark Approval Complete
7
Complete Action
This analyst exhibits 85 distinct sequences across their 147 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 67 | 6.1 | 77.4 | Zero-Edit Visits | 0.21 | 0.00 | 1,359 |
|
Reference SoP
1
Complete Action
Analyst 13's Dominant Sequence
— 8 of 67 instances
1
Continue Workflow
· reference: Complete Action
This analyst exhibits 47 distinct sequences across their 67 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 79 | 7.9 | 77.3 | Zero-Edit Visits | 0.15 | 0.00 | 728 |
|
Reference SoP
1
Complete Action
Analyst 05's Dominant Sequence
— 10 of 79 instances
1
Complete Action
This analyst exhibits 63 distinct sequences across their 79 instances; the dominant one is shown.
|
||||||||
83.5
Avg Adherence Score (0-100)
Median 83.3 · Range 52.6–97.5
47.2 min
Avg Task Duration
0.9 min median
5.3
Avg Sub-Steps per Instance
86.6% have multiple sub-steps
—
Reference Path Coverage
628 sub-task instances
Reference Baseline
Advance Case — Mined SoP
The most common sub-step sequence observed across 628 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Complete Action
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 3,013 | 91.1% | 98.1% |
| Microsoft Teams | 58 | 1.8% | 0.0% |
| Microsoft Excel | 42 | 1.3% | 0.0% |
| Microsoft Outlook | 40 | 1.2% | 0.0% |
| collaboration.merck.com | 36 | 1.1% | 0.0% |
| Acrobat | 28 | 0.8% | 0.0% |
| Microsoft Word | 23 | 0.7% | 0.0% |
| usc-excel.officeapps.live.com | 15 | 0.5% | 0.0% |
| app.powerbi.com | 5 | 0.2% | 0.0% |
| gpteal.merck.com | 5 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Veeva Safety | Microsoft Teams | 42 |
| Veeva Safety | Microsoft Excel | 33 |
| Veeva Safety | Microsoft Outlook | 31 |
| Microsoft Teams | Veeva Safety | 29 |
| Microsoft Outlook | Veeva Safety | 21 |
| Veeva Safety | Acrobat | 21 |
| Veeva Safety | Microsoft Word | 18 |
| Acrobat | Veeva Safety | 18 |
| Veeva Safety | collaboration.merck.com | 15 |
| collaboration.merck.com | usc-excel.officeapps.live.com | 14 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 9,857 events 378 cases 132.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
1941×
pyzeClick
<a>
"Show more"
368×
pyzeClick
<span>
"Narrative"
311×
pyzeClick
<li>
"Narrative"
308×
pyzeClick
<span>
"Approval Complete"
306×
pyzeClick
<span>
"Start Approval Workflow"
252×
pyzeClick
<li>
"Details"
243×
pyzeClick
<span>
"Validation Results (Failures & Warnings)"
199×
Phobos DOM 1,934 events 255 cases 55.5 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
322×
pyzeClick
<button>
"Advance"
197×
pyzeClick
<button>
"Save Comments"
161×
pyzeClick
<button>
"Send to process"
88×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
87×
pyzeClick
<button>
"Send to Medical Review"
67×
pyzeClick
<a>
"Cases"
65×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
63×
Microsoft Word Document 329 events 129 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 528191 (English) (26300681_0_1_338349)
382×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
306×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
294×
file
V-Narrative 542512 (English) (26522671_0_2_344761)
255×
file
V-Narrative 531284 (English) (26355806_0_2_340584)
225×
file
V-Narrative 525091 (English) (26253532_0_1_338428)
201×
file
V-Narrative 531868 (English) (26362311_0_1_340534)
185×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
Microsoft Teams App 217 events 107 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,461 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Acrobat App 214 events 83 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,067 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Outlook Message 127 events 80 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
subj
RE: [Confidential] <Due on 24H> <Approved> <Correction multiple cases>Product Selection
32×
subj
<compensación> Case#2252065
31×
subj
<Due today><Void request><GPVCM POLAND> Case # 2326062
23×
subj
<GPVCM Announcement> Promotions to Safety Data Manager positions in Poland
18×
subj
<compensación> Case#2351734
17×
subj
Due today reports on 03-DEC-2025
14×
subj
Due today reports on 01-DEC-2025
13×
subj
RE: <compensación> Case#2309077
8×
Microsoft Excel Document 105 events 59 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
collaboration.merck.com App 98 events 52 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
ONENOTE Document 41 events 27 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Notepad App 33 events 18 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.07
Swivel Rate
events in non-primary app / total
0.56
Zero-Edit Visit Rate
inspection-only clicks
2.45
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.56
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.56
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 2.45
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.07
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2345589 | Analyst 13 | 52.6 | 6.2m | 25 | 0.68 | 0.00 | 6 |
| 2312686 | Analyst 12 | 56.5 | 33.7m | 10 | 0.70 | 0.00 | 3 |
| 2353949 | Analyst 01 | 57.5 | 4.2m | 4 | 0.75 | 0.00 | 2 |
| 2235982 | Analyst 13 | 57.9 | 2.4m | 21 | 0.52 | 0.00 | 6 |
| 2351742 | Analyst 12 | 60.0 | 0.7m | 6 | 0.67 | 0.00 | 2 |
Users on This Task
13 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
06
01
11
02
15
03
04
14
12
07
13
05
09
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 06 | 2 | 2.0 | 87.5 | Zero-Edit Visits | 0.00 | 0.00 | 9 |
|
Reference SoP
1
Complete Action
Analyst 06's Dominant Sequence
— 1 of 2 instances
1
Complete Action
2
Start Approval Workflow
3
Continue Workflow
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 134 | 8.9 | 85.7 | Zero-Edit Visits | 0.08 | 0.00 | 16,806 |
|
Reference SoP
1
Complete Action
Analyst 01's Dominant Sequence
— 17 of 134 instances
1
Complete Action
2
Mark Approval Complete
3
Complete Action
This analyst exhibits 62 distinct sequences across their 134 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 3 | 3.0 | 85.4 | Zero-Edit Visits | 0.00 | 0.00 | 6 |
|
Reference SoP
1
Complete Action
Analyst 11's Dominant Sequence
— 1 of 3 instances
1
Complete Action
2
Select Send to Case Approval
3
Complete Action
4
Select Start Approval Workflow
5
Complete Action
6
Select Approval Complete
7
Complete Action
8
Navigate to Inbox
This analyst exhibits 3 distinct sequences across their 3 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 112 | 9.3 | 84.5 | Zero-Edit Visits | 0.06 | 0.00 | 215 |
|
Reference SoP
1
Complete Action
Analyst 02's Dominant Sequence
— 23 of 112 instances
1
Complete Action
2
Mark Approval Complete
3
Complete Action
This analyst exhibits 59 distinct sequences across their 112 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 5 | 5.0 | 83.6 | Zero-Edit Visits | 0.00 | 0.00 | 13 |
|
Reference SoP
1
Complete Action
Analyst 15's Dominant Sequence
— 1 of 5 instances
1
Complete Action
2
Send to Case Approval
3
Complete Action
4
Start Approval Workflow
5
Complete Link
6
Mark Approval Complete
7
Complete Action
8
Navigate to Cases
This analyst exhibits 5 distinct sequences across their 5 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 76 | 5.8 | 83.4 | Zero-Edit Visits | 0.07 | 0.00 | 3,069 |
|
Reference SoP
1
Complete Action
Analyst 03's Dominant Sequence
— 14 of 76 instances
1
Complete Action
This analyst exhibits 46 distinct sequences across their 76 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 89 | 6.4 | 83.4 | Zero-Edit Visits | 0.00 | 0.00 | 756 |
|
Reference SoP
1
Complete Action
Analyst 04's Dominant Sequence
— 14 of 89 instances
1
Complete Action
This analyst exhibits 42 distinct sequences across their 89 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 14 | 7.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 38 |
|
Reference SoP
1
Complete Action
Analyst 14's Dominant Sequence
— 2 of 14 instances
1
Complete Action
This analyst exhibits 13 distinct sequences across their 14 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 40 | 5.7 | 82.4 | Zero-Edit Visits | 0.10 | 0.00 | 902 |
|
Reference SoP
1
Complete Action
Analyst 12's Dominant Sequence
— 4 of 40 instances
1
Complete Action
This analyst exhibits 35 distinct sequences across their 40 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 1 | 1.0 | 81.7 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Complete Action
Analyst 07's Dominant Sequence
— 1 of 1 instances
1
Complete Action
2
Select Send to Case Approval
3
Complete Action
4
Select Start Approval Workflow
5
Complete Action
6
Mark Approval Complete
7
View Required Fields Dialog
8
Complete Action
9
Navigate to Cases
|
||||||||
| ▸ | Analyst 13 | 38 | 3.8 | 81.4 | Zero-Edit Visits | 0.14 | 0.00 | 77 |
|
Reference SoP
1
Complete Action
Analyst 13's Dominant Sequence
— 5 of 38 instances
1
Complete Action
This analyst exhibits 28 distinct sequences across their 38 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 44 | 4.4 | 81.2 | Zero-Edit Visits | 0.06 | 0.00 | 433 |
|
Reference SoP
1
Complete Action
Analyst 05's Dominant Sequence
— 10 of 44 instances
1
Complete Action
This analyst exhibits 33 distinct sequences across their 44 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 70 | 4.4 | 81.1 | Zero-Edit Visits | 0.10 | 0.00 | 7,320 |
|
Reference SoP
1
Complete Action
Analyst 09's Dominant Sequence
— 17 of 70 instances
1
Complete Action
2
Send to Case Approval
3
Complete Action
4
Start Approval Workflow
5
Complete Action
6
Mark Approval Complete
7
Complete Action
This analyst exhibits 34 distinct sequences across their 70 instances; the dominant one is shown.
|
||||||||
78.5
Avg Adherence Score (0-100)
Median 82.5 · Range 40.2–97.5
40.3 min
Avg Task Duration
1.5 min median
6.2
Avg Sub-Steps per Instance
68.1% have multiple sub-steps
—
Reference Path Coverage
210 sub-task instances
Reference Baseline
Advance Case — Mined SoP
The most common sub-step sequence observed across 210 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Continue Workflow
2
Delete Item
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 1,098 | 84.1% | 99.0% |
| Microsoft Word | 56 | 4.3% | 0.0% |
| Acrobat | 53 | 4.1% | 0.0% |
| Microsoft Teams | 29 | 2.2% | 0.0% |
| collaboration.merck.com | 13 | 1.0% | 0.0% |
| Microsoft Outlook | 12 | 0.9% | 0.0% |
| Microsoft Excel | 8 | 0.6% | 0.0% |
| gpteal.merck.com | 6 | 0.5% | 0.0% |
| Notepad | 3 | 0.2% | 0.0% |
| SnippingTool | 3 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Acrobat | Microsoft Word | 28 |
| Microsoft Word | Acrobat | 26 |
| Veeva Safety | Acrobat | 24 |
| Veeva Safety | Microsoft Teams | 20 |
| Acrobat | Veeva Safety | 19 |
| Veeva Safety | Microsoft Word | 19 |
| Microsoft Word | Veeva Safety | 15 |
| Microsoft Teams | Veeva Safety | 11 |
| Veeva Safety | Microsoft Outlook | 9 |
| Veeva Safety | Microsoft Excel | 5 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 4,654 events 167 cases 77.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Continue"
447×
pyzeClick
<span>
"Complete"
327×
pyzeClick
<li>
"Delete"
300×
pyzeClick
<button>
"Create"
151×
pyzeClick
<a>
"Show more"
118×
pyzeClick
<li>
"Narrative"
112×
pyzeClick
<span>
"Narrative"
103×
pyzeClick
<li>
"Details"
92×
Phobos DOM 305 events 57 cases 11.0 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
57×
pyzeClick
<button>
"Advance"
33×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
24×
pyzeClick
<button>
"Save Comments"
18×
pyzeClick
<a>
"Cases"
17×
pyzeClick
<button>
">> Send to Quality Check"
13×
pyzeClick
<a>
"Cases Coordinator"
11×
pyzeClick
<button>
"Send to process"
10×
Microsoft Word Document 189 events 48 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 552979 (English) (26689105_0_2_348832)
557×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
515×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
429×
file
V-Narrative 553023 (English) (26689066_0_2_351253)
251×
file
V-Narrative 533882 (English) (26397511_0_1_345751)
230×
file
V-Narrative 549428 (English) (26642399_0_2_349187)
170×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
file
V-Narrative 538518 (English) (26459147_0_1_344628)
169×
Acrobat App 154 events 38 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,365 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 109 events 50 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,019 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Excel Document 54 events 24 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
1761620 (v2.1)
330×
file
2352822 (v0.1)
216×
file
Case tracking list
168×
file
MERCK2516915_ARGPRO_20251125054618
97×
file
Daily tracking list (PV-SMART) copy
87×
file
Priorización
68×
file
ICHICSR-EVHUMAN_10020432161 (1)
30×
file
GDMS800-039-001 GPVCM Finding Definitions (1)
20×
Microsoft Outlook Message 51 events 34 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
collaboration.merck.com App 36 events 21 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Notepad App 19 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
gpteal.merck.com DOM 18 events 10 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.11
Swivel Rate
events in non-primary app / total
0.88
Zero-Edit Visit Rate
inspection-only clicks
2.18
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.88
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.88
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 2.18
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.11
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2309351 | Analyst 01 | 40.2 | 17.0m | 28 | 0.75 | 0.00 | 11 |
| 2341558 | Analyst 13 | 50.8 | 1041.4m | 27 | 0.59 | 0.00 | 7 |
| 2252065 | Analyst 05 | 53.6 | 6.7m | 21 | 0.57 | 0.00 | 6 |
| 2350507 | Analyst 09 | 53.8 | 14.7m | 16 | 0.69 | 0.00 | 5 |
| 2314923 | Analyst 05 | 54.2 | 20.6m | 26 | 0.46 | 0.00 | 7 |
Users on This Task
12 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
04
10
15
07
01
02
03
14
12
13
05
09
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 04 | 42 | 4.2 | 82.9 | Zero-Edit Visits | 0.00 | 0.00 | 642 |
|
Reference SoP
1
Continue Workflow
2
Delete Item
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 04's Dominant Sequence
— 15 of 42 instances
1
Continue Workflow
This analyst exhibits 26 distinct sequences across their 42 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Continue Workflow
2
Delete Item
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 10's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
|
||||||||
| ▸ | Analyst 15 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Continue Workflow
2
Delete Item
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 15's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
2
View Details
· reference: Delete Item
|
||||||||
| ▸ | Analyst 07 | 1 | 1.0 | 80.0 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Continue Workflow
2
Delete Item
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 07's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
2
Show More Details
· reference: Delete Item
3
Select Reference Numbers
· reference: Google Search
4
Delete Item
· reference: pubmed.ncbi.nlm.nih.gov
5
Continue Workflow
· reference: Google Search
6
View Validation Results
· reference: Navigate to Cases
7
Select Products
· reference: Create New Item
|
||||||||
| ▸ | Analyst 01 | 32 | 2.3 | 79.8 | Zero-Edit Visits | 0.09 | 0.00 | 132 |
|
Reference SoP
1
Continue Workflow
2
Delete Item
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 01's Dominant Sequence
— 9 of 32 instances
1
Continue Workflow
This analyst exhibits 22 distinct sequences across their 32 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 29 | 2.9 | 79.5 | Zero-Edit Visits | 0.12 | 0.01 | 3,322 |
|
Reference SoP
1
Continue Workflow
2
Delete Item
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 02's Dominant Sequence
— 7 of 29 instances
1
Continue Workflow
This analyst exhibits 23 distinct sequences across their 29 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 31 | 2.6 | 78.5 | Zero-Edit Visits | 0.10 | 0.00 | 1,266 |
|
Reference SoP
1
Continue Workflow
2
Delete Item
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 03's Dominant Sequence
— 14 of 31 instances
1
Continue Workflow
This analyst exhibits 17 distinct sequences across their 31 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 1 | 1.0 | 75.9 | Zero-Edit Visits | 0.00 | 0.00 | 10 |
|
Reference SoP
1
Continue Workflow
2
Delete Item
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 14's Dominant Sequence
— 1 of 1 instances
1
Continue Workflow
2
Press Enter
· reference: Delete Item
3
Continue Link
· reference: Google Search
4
Select List Item
· reference: pubmed.ncbi.nlm.nih.gov
5
View Study
· reference: Google Search
6
Select Change Type
· reference: Navigate to Cases
7
View External Product
· reference: Create New Item
8
Continue Workflow
· reference: Select External Product
9
Select Change Type
· reference: Continue
10
View External Product
· reference: Select Products
11
Continue Workflow
· reference: Delete Item
12
Select Change Type
· reference: Continue Workflow
13
View External Product
· reference: Notepad
14
Continue Link
15
Continue Workflow
16
View Products
17
Select Change Type
18
View External Product
19
Continue Workflow
20
View Case Assessments
21
Select Not Applicable
22
View Case Assessment Results
|
||||||||
| ▸ | Analyst 12 | 14 | 2.8 | 75.8 | Zero-Edit Visits | 0.19 | 0.00 | 188 |
|
Reference SoP
1
Continue Workflow
2
Delete Item
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 12's Dominant Sequence
— 2 of 14 instances
1
Continue Workflow
This analyst exhibits 13 distinct sequences across their 14 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 16 | 1.8 | 75.6 | Zero-Edit Visits | 0.21 | 0.00 | 1,209 |
|
Reference SoP
1
Continue Workflow
2
Delete Item
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 13's Dominant Sequence
— 8 of 16 instances
1
Continue Workflow
This analyst exhibits 9 distinct sequences across their 16 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 15 | 2.1 | 75.6 | Zero-Edit Visits | 0.21 | 0.00 | 88 |
|
Reference SoP
1
Continue Workflow
2
Delete Item
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 05's Dominant Sequence
— 2 of 15 instances
1
Continue Workflow
This analyst exhibits 14 distinct sequences across their 15 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 27 | 2.2 | 73.3 | Zero-Edit Visits | 0.17 | 0.00 | 1,602 |
|
Reference SoP
1
Continue Workflow
2
Delete Item
3
Google Search
4
pubmed.ncbi.nlm.nih.gov
5
Google Search
6
Navigate to Cases
7
Create New Item
8
Select External Product
9
Continue
10
Select Products
11
Delete Item
12
Continue Workflow
13
Notepad
Analyst 09's Dominant Sequence
— 9 of 27 instances
1
Continue Workflow
This analyst exhibits 19 distinct sequences across their 27 instances; the dominant one is shown.
|
||||||||
79.0
Avg Adherence Score (0-100)
Median 82.5 · Range 43.4–97.5
5.6 min
Avg Task Duration
0.9 min median
4.7
Avg Sub-Steps per Instance
65.9% have multiple sub-steps
—
Reference Path Coverage
346 sub-task instances
Reference Baseline
Advance Case — Mined SoP
The most common sub-step sequence observed across 346 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Save Changes
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Veeva Safety | 1,296 | 79.1% | 100.0% |
| Microsoft Word | 81 | 4.9% | 0.0% |
| Acrobat | 65 | 4.0% | 0.0% |
| Microsoft Teams | 44 | 2.7% | 0.0% |
| collaboration.merck.com | 39 | 2.4% | 0.0% |
| Microsoft Excel | 30 | 1.8% | 0.0% |
| Microsoft Outlook | 24 | 1.5% | 0.0% |
| usc-excel.officeapps.live.com | 12 | 0.7% | 0.0% |
| Notepad | 5 | 0.3% | 0.0% |
| ONENOTE | 5 | 0.3% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Veeva Safety | Microsoft Word | 40 |
| Veeva Safety | Acrobat | 34 |
| Acrobat | Veeva Safety | 28 |
| Microsoft Word | Veeva Safety | 27 |
| Veeva Safety | Microsoft Teams | 25 |
| Microsoft Word | Acrobat | 21 |
| Acrobat | Microsoft Word | 21 |
| Microsoft Teams | Veeva Safety | 20 |
| Veeva Safety | collaboration.merck.com | 20 |
| Veeva Safety | Microsoft Outlook | 17 |
On-Case Application Journey (±15 min around anchor)
Apps the same analyst touched on the same case within ±15 minutes of the task anchor. Captures work that spans task boundaries — the Veeva/Acrobat/Outlook/Word context surrounding the anchor click.
Case counts per app are not mutually exclusive — a case that touched Veeva AND Phobos in this window is counted in both rows. Overlap reflects real cross-app work, not double-counted cases.
Veeva Safety DOM 6,667 events 254 cases 125.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
449×
pyzeClick
<button>
"Save"
422×
pyzeClick
<span>
"Narrative"
217×
pyzeClick
<li>
"Narrative"
183×
pyzeClick
<a>
"Show more"
183×
pyzeClick
<span>
"Adverse Events"
179×
pyzeClick
<li>
"Details"
149×
pyzeClick
<button>
"Create"
133×
Phobos DOM 754 events 111 cases 26.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
105×
pyzeClick
<button>
"Advance"
71×
pyzeClick
<button>
"Save Comments"
59×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
52×
pyzeClick
<button>
"Send to process"
40×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
24×
pyzeClick
<a>
"Cases"
22×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
19×
Microsoft Word Document 281 events 81 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 552979 (English) (26689105_0_2_348832)
1076×
file
V-Narrative 549550 (English) (26643764_0_1_347593)
526×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
515×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
377×
file
V-Narrative 533909 (English) (26397745_0_2_338171)
243×
file
V-Narrative 545329 (English) (26567747_0_2_344796)
215×
file
V-Narrative 540358 (English) (26494608_0_2_342069)
195×
file
Narrative 526208 (English)
190×
Acrobat App 207 events 57 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
3,088 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 163 events 68 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,132 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Excel Document 110 events 44 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Case tracking list
382×
file
2352822 (v0.1)
234×
file
Daily tracking list (PV-SMART) copy
210×
file
I-0000353124
142×
file
I-0000350382
141×
file
Priorización
61×
file
MERCK2524493_ARGPRO_20251202045619
46×
file
ICHICSR-EVHUMAN_10020432161 (1)
37×
collaboration.merck.com App 100 events 48 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Microsoft Outlook Message 83 events 49 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
usc-excel.officeapps.live.com Document 24 events 6 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
gpteal.merck.com DOM 22 events 11 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.13
Swivel Rate
events in non-primary app / total
0.89
Zero-Edit Visit Rate
inspection-only clicks
1.56
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.89
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Vector Contribution to Adherence Loss
Each vector's weighted contribution to lost adherence. The top row is your highest-leverage fix.
Zero-Edit Visit Rate
value 0.89
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.56
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.13
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 1677926 | Analyst 05 | 43.4 | 34.9m | 48 | 0.58 | 0.00 | 14 |
| 2353221 | Analyst 13 | 48.3 | 0.3m | 9 | 0.89 | 0.00 | 4 |
| 2329230 | Analyst 05 | 48.8 | 2.9m | 8 | 0.88 | 0.00 | 4 |
| 2341841 | Analyst 13 | 50.8 | 10.2m | 25 | 0.68 | 0.00 | 6 |
| 2329808 | Analyst 13 | 51.0 | 6.9m | 10 | 0.80 | 0.00 | 4 |
Users on This Task
15 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
06
15
04
14
16
10
11
07
01
03
02
09
12
05
13
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 06 | 1 | 1.0 | 90.0 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Save Changes
Analyst 06's Dominant Sequence
— 1 of 1 instances
1
Save Changes
2
Select List Item
3
Save Changes
|
||||||||
| ▸ | Analyst 15 | 4 | 4.0 | 86.1 | Zero-Edit Visits | 0.00 | 0.00 | 26 |
|
Reference SoP
1
Save Changes
Analyst 15's Dominant Sequence
— 2 of 4 instances
1
Save Changes
This analyst exhibits 3 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 70 | 5.8 | 85.2 | Zero-Edit Visits | 0.00 | 0.00 | 197 |
|
Reference SoP
1
Save Changes
Analyst 04's Dominant Sequence
— 39 of 70 instances
1
Save Changes
This analyst exhibits 26 distinct sequences across their 70 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 11 | 5.5 | 82.6 | Zero-Edit Visits | 0.00 | 0.00 | 41 |
|
Reference SoP
1
Save Changes
Analyst 14's Dominant Sequence
— 4 of 11 instances
1
Save Changes
This analyst exhibits 8 distinct sequences across their 11 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 16 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 8 |
|
Reference SoP
1
Save Changes
Analyst 16's Dominant Sequence
— 1 of 1 instances
1
Save Changes
2
Complete Action
|
||||||||
| ▸ | Analyst 10 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Save Changes
Analyst 10's Dominant Sequence
— 1 of 1 instances
1
Save Changes
|
||||||||
| ▸ | Analyst 11 | 4 | 4.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 16 |
|
Reference SoP
1
Save Changes
Analyst 11's Dominant Sequence
— 3 of 4 instances
1
Save Changes
This analyst exhibits 2 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 07 | 4 | 4.0 | 81.6 | Zero-Edit Visits | 0.00 | 0.00 | 12 |
|
Reference SoP
1
Save Changes
Analyst 07's Dominant Sequence
— 2 of 4 instances
1
Save Changes
This analyst exhibits 3 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 35 | 2.5 | 80.6 | Zero-Edit Visits | 0.11 | 0.00 | 108 |
|
Reference SoP
1
Save Changes
Analyst 01's Dominant Sequence
— 7 of 35 instances
1
Save Changes
This analyst exhibits 29 distinct sequences across their 35 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 41 | 3.4 | 80.4 | Zero-Edit Visits | 0.09 | 0.00 | 79 |
|
Reference SoP
1
Save Changes
Analyst 03's Dominant Sequence
— 16 of 41 instances
1
Save Changes
This analyst exhibits 24 distinct sequences across their 41 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 36 | 3.3 | 77.1 | Zero-Edit Visits | 0.15 | 0.00 | 139 |
|
Reference SoP
1
Save Changes
Analyst 02's Dominant Sequence
— 6 of 36 instances
1
Save Changes
This analyst exhibits 29 distinct sequences across their 36 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 50 | 3.6 | 76.7 | Zero-Edit Visits | 0.18 | 0.00 | 236 |
|
Reference SoP
1
Save Changes
Analyst 09's Dominant Sequence
— 15 of 50 instances
1
Save Changes
This analyst exhibits 32 distinct sequences across their 50 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 55 | 6.1 | 76.5 | Zero-Edit Visits | 0.20 | 0.00 | 786 |
|
Reference SoP
1
Save Changes
Analyst 12's Dominant Sequence
— 14 of 55 instances
1
Save Changes
This analyst exhibits 37 distinct sequences across their 55 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 20 | 2.2 | 69.8 | Zero-Edit Visits | 0.30 | 0.00 | 207 |
|
Reference SoP
1
Save Changes
Analyst 05's Dominant Sequence
— 5 of 20 instances
1
Save Changes
This analyst exhibits 16 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 13 | 1.9 | 68.0 | Zero-Edit Visits | 0.39 | 0.00 | 72 |
|
Reference SoP
1
Save Changes
Analyst 13's Dominant Sequence
— 4 of 13 instances
1
Save Changes
This analyst exhibits 10 distinct sequences across their 13 instances; the dominant one is shown.
|
||||||||