Task-Boundary Mode
How task-instance boundaries are drawn from the event stream. Applies to every Task SoP, Step SoP, and Variants view.
Narrative
📝 Task SoP
Mined from 595 instances
Select Narrative Template
How 14 analysts execute this task across 318 cases — auto-generated Standard Operating Procedure with adherence scoring.
80.0
Avg Adherence Score (0-100)
Median 82.5 · Range 33.1–97.5
1.9 min
Avg Task Duration
0.4 min median
3.3
Avg Sub-Steps per Instance
75.1% have multiple sub-steps
17.1%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
17.1% of instances follow this exactly
Select Narrative Template — Mined SoP
The most common sub-step sequence observed across 651 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Select Narrative
0.6s
0.1 edits
Veeva Safety
100%
2
Show More Details
0.3s
0.1 edits
Veeva Safety
44%
Variant Explorer
Reference
Select Narrative Template Reference Path
111 instances · 9 analysts
This is the mined reference path — 111 instances across 9 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Select Narrative Template.
Genuine Complexity
Complex-Case Execution Path
162 instances · 11 analysts
· Δ adherence +4.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
×
Show More Details
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 84.3 vs reference 80.0 (delta +4.3). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Reviewer Sweep
Reviewer Verification Pass
7 instances · 2 analysts
· Δ adherence +2.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
View Migration Notice
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 2 analysts across 6 cases.
Genuine Complexity
Complex-Case Execution Path
7 instances · 2 analysts
· Δ adherence +2.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
View Validation Results
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 82.5 vs reference 80.0 (delta +2.5). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
6 instances · 3 analysts
· Δ adherence +12.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Preview Japanese Narrative
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 92.5 vs reference 80.0 (delta +12.5). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Word
6 instances · 4 analysts
· Δ adherence -13.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Microsoft Word
reference at this step: Show More Details
Cross-App Swivel — 4 analysts pivot into Microsoft Word during this task (0% of task time in non-primary apps, 53% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Genuine Complexity
Complex-Case Execution Path
5 instances · 2 analysts
· Δ adherence +4.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Select Documents
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 84.0 vs reference 80.0 (delta +4.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Word
5 instances · 3 analysts
· Δ adherence -9.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
Microsoft Word
extra step (reference ends at step 2)
Cross-App Swivel — 3 analysts pivot into Microsoft Word during this task (0% of task time in non-primary apps, 40% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Genuine Complexity
Complex-Case Execution Path
5 instances · 2 analysts
· Δ adherence +7.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Preview English Narrative
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 87.0 vs reference 80.0 (delta +7.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Reviewer Sweep
Reviewer Verification Pass
4 instances · 1 analysts
· Δ adherence +2.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
View Physician Report
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 1 analysts across 3 cases.
Cross-App Swivel
Cross-App Lookup to Microsoft Teams
4 instances · 2 analysts
· Δ adherence -7.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
Microsoft Teams
extra step (reference ends at step 2)
Cross-App Swivel — 2 analysts pivot into Microsoft Teams during this task (0% of task time in non-primary apps, 33% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Reviewer Sweep
Reviewer Verification Pass
4 instances · 2 analysts
· Δ adherence +2.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
Select Adverse Events
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 2 analysts across 4 cases.
Genuine Complexity
Complex-Case Execution Path
4 instances · 2 analysts
· Δ adherence +4.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Select Submissions
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 84.4 vs reference 80.0 (delta +4.4). 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 | 1,702 | 78.7% | 87.1% |
| Microsoft Word | 152 | 7.0% | 0.0% |
| Acrobat | 99 | 4.6% | 0.0% |
| Microsoft Teams | 53 | 2.4% | 0.0% |
| gpteal.merck.com | 31 | 1.4% | 0.0% |
| Notepad | 22 | 1.0% | 0.0% |
| collaboration.merck.com | 19 | 0.9% | 0.0% |
| Microsoft Excel | 17 | 0.8% | 0.0% |
| Microsoft Outlook | 14 | 0.6% | 0.0% |
| Quality Vault | 6 | 0.3% | 12.9% |
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 | 67 |
| Microsoft Word | Acrobat | 57 |
| Acrobat | Microsoft Word | 52 |
| Microsoft Word | Veeva Safety | 38 |
| Veeva Safety | Acrobat | 31 |
| Veeva Safety | Microsoft Teams | 31 |
| Veeva Safety | gpteal.merck.com | 22 |
| Acrobat | Veeva Safety | 19 |
| Veeva Safety | Notepad | 17 |
| Veeva Safety | collaboration.merck.com | 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 7,023 events 318 cases 70.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<li>
"Narrative"
683×
pyzeClick
<span>
"Complete"
593×
pyzeClick
<a>
"Show more"
520×
pyzeClick
<li>
"Details"
288×
pyzeClick
<button>
"Create"
131×
pyzeClick
<li>
"Patient"
122×
pyzeClick
<span>
"Validation Results (Failures & Warnings)"
117×
pyzeClick
<button>
"Save"
115×
Phobos DOM 1,104 events 155 cases 30.0 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
153×
pyzeClick
<button>
"Advance"
89×
pyzeClick
<button>
"Save Comments"
77×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
52×
pyzeClick
<button>
"Send to process"
43×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
40×
pyzeClick
<a>
"Cases"
35×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
34×
Microsoft Word Document 441 events 128 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 540591 (English) (26495494_0_2_345212)
726×
file
V-Narrative 533164 (English) (26378749_0_4_342805)
659×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
572×
file
V-Narrative 553016 (English) (26689302_0_2_348778)
392×
file
V-Narrative 530932 (English) (26349278_0_2_339374)
387×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
382×
file
V-Narrative 533533 (English) (26384397_0_2_337752)
361×
file
V-Narrative 525025 (English) (26253159_0_3_342205)
320×
Acrobat App 289 events 71 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,900 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 192 events 87 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,339 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 100 events 52 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Daily tracking list (PV-SMART) copy
523×
file
E2B_MSDPRO_014172_AGPROD1_20251127163042
155×
file
Priorización
152×
file
I-0000350382
142×
file
records_as_of_2025_11_28_EST
140×
file
MERCK2516915_ARGPRO_20251125054618
111×
file
Case tracking list
82×
file
MERCK2524493_ARGPRO_20251202045619
46×
Microsoft Outlook Message 72 events 44 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
collaboration.merck.com App 68 events 37 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
gpteal.merck.com DOM 56 events 33 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Notepad App 54 events 30 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.
0.11
Swivel Rate
events in non-primary app / total
0.92
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 0.92
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.
Swivel Rate
value 0.11
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 |
|---|---|---|---|---|---|---|---|
| 2335834 | Analyst 01 | 33.1 | 8.2m | 29 | 0.90 | 0.00 | 13 |
| 2290475 | Analyst 01 | 33.2 | 6.5m | 28 | 0.89 | 0.00 | 13 |
| 2350181 | Analyst 01 | 38.9 | 26.2m | 32 | 0.72 | 0.00 | 11 |
| 2309351 | Analyst 01 | 46.0 | 14.3m | 15 | 0.80 | 0.00 | 6 |
| 2351004 | Analyst 01 | 50.0 | 5.1m | 12 | 0.75 | 0.00 | 5 |
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) →
15
11
07
04
14
08
03
12
01
02
09
05
10
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 15 | 5 | 5.0 | 90.3 | Zero-Edit Visits | 0.00 | 0.00 | 12 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 15's Dominant Sequence
— 4 of 5 instances
1
Select Narrative
This analyst exhibits 2 distinct sequences across their 5 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 3 | 3.0 | 87.5 | Zero-Edit Visits | 0.00 | 0.00 | 3 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 11's Dominant Sequence
— 3 of 3 instances
1
Select Narrative
|
||||||||
| ▸ | Analyst 07 | 1 | 1.0 | 85.5 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 07's Dominant Sequence
— 1 of 1 instances
1
Select Narrative
2
Show More Details
3
View Narrative Preview Header
4
Update Case Narrative
5
Select List Item
|
||||||||
| ▸ | Analyst 04 | 6 | 1.2 | 85.0 | Zero-Edit Visits | 0.00 | 0.00 | 19 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 04's Dominant Sequence
— 4 of 6 instances
1
Select Narrative
This analyst exhibits 3 distinct sequences across their 6 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 9 | 4.5 | 83.8 | Zero-Edit Visits | 0.00 | 0.00 | 20 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 14's Dominant Sequence
— 4 of 9 instances
1
Select Narrative
This analyst exhibits 6 distinct sequences across their 9 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 08 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 08's Dominant Sequence
— 1 of 1 instances
1
Select Narrative
2
Show More Details
|
||||||||
| ▸ | Analyst 03 | 20 | 2.0 | 82.0 | Zero-Edit Visits | 0.02 | 0.00 | 11 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 03's Dominant Sequence
— 10 of 20 instances
1
Select Narrative
This analyst exhibits 7 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 31 | 4.4 | 80.4 | Zero-Edit Visits | 0.14 | 0.00 | 101 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 12's Dominant Sequence
— 8 of 31 instances
1
Select Narrative
This analyst exhibits 24 distinct sequences across their 31 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 238 | 15.9 | 80.3 | Zero-Edit Visits | 0.10 | 0.00 | 480 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 01's Dominant Sequence
— 66 of 238 instances
1
Select Narrative
This analyst exhibits 121 distinct sequences across their 238 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 240 | 20.0 | 80.3 | Zero-Edit Visits | 0.09 | 0.00 | 263 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 02's Dominant Sequence
— 52 of 240 instances
1
Select Narrative
2
Show More Details
This analyst exhibits 110 distinct sequences across their 240 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 20 | 2.0 | 80.1 | Zero-Edit Visits | 0.07 | 0.00 | 68 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 09's Dominant Sequence
— 6 of 20 instances
1
Select Narrative
This analyst exhibits 11 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 6 | 1.2 | 77.9 | Zero-Edit Visits | 0.25 | 0.00 | 4 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 05's Dominant Sequence
— 1 of 6 instances
1
Select Narrative
2
Acrobat
· reference: Show More Details
3
Microsoft Word
This analyst exhibits 6 distinct sequences across their 6 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 49 | 4.5 | 76.0 | Zero-Edit Visits | 0.18 | 0.00 | 171 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 10's Dominant Sequence
— 9 of 49 instances
1
Select Narrative
2
Show More Details
This analyst exhibits 32 distinct sequences across their 49 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 22 | 2.4 | 73.6 | Zero-Edit Visits | 0.31 | 0.00 | 65 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 13's Dominant Sequence
— 4 of 22 instances
1
Select Narrative
This analyst exhibits 18 distinct sequences across their 22 instances; the dominant one is shown.
|
||||||||
79.9
Avg Adherence Score (0-100)
Median 82.5 · Range 33.1–97.5
2.3 min
Avg Task Duration
0.4 min median
3.5
Avg Sub-Steps per Instance
75.4% have multiple sub-steps
16.9%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
16.9% of instances follow this exactly
Select Narrative Template — Mined SoP
The most common sub-step sequence observed across 651 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Select Narrative
0.6s
0.1 edits
Veeva Safety
100%
2
Show More Details
0.3s
0.1 edits
Veeva Safety
43%
Variant Explorer
Reference
Select Narrative Template Reference Path
110 instances · 9 analysts
This is the mined reference path — 110 instances across 9 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Select Narrative Template.
Genuine Complexity
Complex-Case Execution Path
160 instances · 11 analysts
· Δ adherence +4.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
×
Show More Details
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 84.3 vs reference 79.9 (delta +4.4). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
7 instances · 2 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
View Validation Results
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 82.5 vs reference 79.9 (delta +2.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
6 instances · 3 analysts
· Δ adherence +12.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Preview Japanese Narrative
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 92.5 vs reference 79.9 (delta +12.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Reviewer Sweep
Reviewer Verification Pass
5 instances · 2 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
View Migration Notice
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 2 analysts across 5 cases.
Genuine Complexity
Complex-Case Execution Path
5 instances · 2 analysts
· Δ adherence +7.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Preview English Narrative
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 87.0 vs reference 79.9 (delta +7.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Word
5 instances · 4 analysts
· Δ adherence -13.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Microsoft Word
reference at this step: Show More Details
Cross-App Swivel — 4 analysts pivot into Microsoft Word during this task (0% of task time in non-primary apps, 53% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Genuine Complexity
Complex-Case Execution Path
4 instances · 2 analysts
· Δ adherence +4.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Select Documents
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 84.4 vs reference 79.9 (delta +4.5). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Teams
4 instances · 2 analysts
· Δ adherence -7.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
Microsoft Teams
extra step (reference ends at step 2)
Cross-App Swivel — 2 analysts pivot into Microsoft Teams during this task (0% of task time in non-primary apps, 33% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Reviewer Sweep
Reviewer Verification Pass
4 instances · 2 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
Select Adverse Events
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 2 analysts across 4 cases.
Genuine Complexity
Complex-Case Execution Path
4 instances · 2 analysts
· Δ adherence +4.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Select Submissions
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 84.4 vs reference 79.9 (delta +4.5). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Acrobat
4 instances · 1 analysts
· Δ adherence -7.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
Acrobat
extra step (reference ends at step 2)
Cross-App Swivel — 1 analysts pivot into Acrobat during this task (0% of task time in non-primary apps, 33% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Reviewer Sweep
Reviewer Verification Pass
4 instances · 2 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
View Investigator Information
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 2 analysts across 3 cases.
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 | 1,776 | 78.4% | 87.5% |
| Microsoft Word | 165 | 7.3% | 0.0% |
| Acrobat | 101 | 4.5% | 0.0% |
| Microsoft Teams | 56 | 2.5% | 0.0% |
| gpteal.merck.com | 33 | 1.5% | 0.0% |
| Notepad | 22 | 1.0% | 0.0% |
| collaboration.merck.com | 21 | 0.9% | 0.0% |
| Microsoft Excel | 19 | 0.8% | 0.0% |
| Microsoft Outlook | 14 | 0.6% | 0.0% |
| dtgpteal.merck.com | 6 | 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 | 72 |
| Microsoft Word | Acrobat | 57 |
| Acrobat | Microsoft Word | 50 |
| Microsoft Word | Veeva Safety | 48 |
| Veeva Safety | Microsoft Teams | 33 |
| Veeva Safety | Acrobat | 31 |
| Acrobat | Veeva Safety | 23 |
| Veeva Safety | gpteal.merck.com | 23 |
| Veeva Safety | Notepad | 17 |
| Microsoft Teams | Veeva Safety | 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 7,023 events 318 cases 70.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<li>
"Narrative"
683×
pyzeClick
<span>
"Complete"
593×
pyzeClick
<a>
"Show more"
520×
pyzeClick
<li>
"Details"
288×
pyzeClick
<button>
"Create"
131×
pyzeClick
<li>
"Patient"
122×
pyzeClick
<span>
"Validation Results (Failures & Warnings)"
117×
pyzeClick
<button>
"Save"
115×
Phobos DOM 1,104 events 155 cases 30.0 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
153×
pyzeClick
<button>
"Advance"
89×
pyzeClick
<button>
"Save Comments"
77×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
52×
pyzeClick
<button>
"Send to process"
43×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
40×
pyzeClick
<a>
"Cases"
35×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
34×
Microsoft Word Document 441 events 128 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 540591 (English) (26495494_0_2_345212)
726×
file
V-Narrative 533164 (English) (26378749_0_4_342805)
659×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
572×
file
V-Narrative 553016 (English) (26689302_0_2_348778)
392×
file
V-Narrative 530932 (English) (26349278_0_2_339374)
387×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
382×
file
V-Narrative 533533 (English) (26384397_0_2_337752)
361×
file
V-Narrative 525025 (English) (26253159_0_3_342205)
320×
Acrobat App 289 events 71 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,900 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 192 events 87 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,339 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 100 events 52 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Daily tracking list (PV-SMART) copy
523×
file
E2B_MSDPRO_014172_AGPROD1_20251127163042
155×
file
Priorización
152×
file
I-0000350382
142×
file
records_as_of_2025_11_28_EST
140×
file
MERCK2516915_ARGPRO_20251125054618
111×
file
Case tracking list
82×
file
MERCK2524493_ARGPRO_20251202045619
46×
Microsoft Outlook Message 72 events 44 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
collaboration.merck.com App 68 events 37 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
gpteal.merck.com DOM 56 events 33 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Notepad App 54 events 30 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.
0.11
Swivel Rate
events in non-primary app / total
0.92
Zero-Edit Visit Rate
inspection-only clicks
1.24
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.
Swivel Rate
value 0.11
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Revisit Depth
value 1.24
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 |
|---|---|---|---|---|---|---|---|
| 2335834 | Analyst 01 | 33.1 | 8.2m | 29 | 0.90 | 0.00 | 13 |
| 2290475 | Analyst 01 | 33.2 | 6.5m | 28 | 0.89 | 0.00 | 13 |
| 2350181 | Analyst 01 | 38.9 | 26.2m | 32 | 0.72 | 0.00 | 11 |
| 2309351 | Analyst 01 | 47.8 | 15.9m | 19 | 0.68 | 0.00 | 7 |
| 2351004 | Analyst 01 | 50.0 | 5.1m | 12 | 0.75 | 0.00 | 5 |
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) →
15
11
07
04
14
08
03
01
02
09
12
05
10
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 15 | 5 | 5.0 | 90.3 | Zero-Edit Visits | 0.00 | 0.00 | 12 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 15's Dominant Sequence
— 4 of 5 instances
1
Select Narrative
This analyst exhibits 2 distinct sequences across their 5 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 3 | 3.0 | 87.5 | Zero-Edit Visits | 0.00 | 0.00 | 3 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 11's Dominant Sequence
— 3 of 3 instances
1
Select Narrative
|
||||||||
| ▸ | Analyst 07 | 1 | 1.0 | 85.5 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 07's Dominant Sequence
— 1 of 1 instances
1
Select Narrative
2
Show More Details
3
View Narrative Preview Header
4
Update Case Narrative
5
Select List Item
|
||||||||
| ▸ | Analyst 04 | 6 | 1.2 | 85.0 | Zero-Edit Visits | 0.00 | 0.00 | 19 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 04's Dominant Sequence
— 4 of 6 instances
1
Select Narrative
This analyst exhibits 3 distinct sequences across their 6 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 9 | 4.5 | 83.8 | Zero-Edit Visits | 0.00 | 0.00 | 29 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 14's Dominant Sequence
— 3 of 9 instances
1
Select Narrative
This analyst exhibits 7 distinct sequences across their 9 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 08 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 08's Dominant Sequence
— 1 of 1 instances
1
Select Narrative
2
Show More Details
|
||||||||
| ▸ | Analyst 03 | 20 | 2.0 | 82.0 | Zero-Edit Visits | 0.02 | 0.00 | 11 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 03's Dominant Sequence
— 10 of 20 instances
1
Select Narrative
This analyst exhibits 7 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 238 | 15.9 | 80.3 | Zero-Edit Visits | 0.10 | 0.00 | 558 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 01's Dominant Sequence
— 66 of 238 instances
1
Select Narrative
This analyst exhibits 124 distinct sequences across their 238 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 240 | 20.0 | 80.3 | Zero-Edit Visits | 0.09 | 0.00 | 322 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 02's Dominant Sequence
— 52 of 240 instances
1
Select Narrative
2
Show More Details
This analyst exhibits 114 distinct sequences across their 240 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 20 | 2.0 | 80.1 | Zero-Edit Visits | 0.07 | 0.00 | 68 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 09's Dominant Sequence
— 6 of 20 instances
1
Select Narrative
This analyst exhibits 11 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 31 | 4.4 | 79.9 | Zero-Edit Visits | 0.15 | 0.00 | 132 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 12's Dominant Sequence
— 7 of 31 instances
1
Select Narrative
This analyst exhibits 25 distinct sequences across their 31 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 6 | 1.2 | 77.9 | Zero-Edit Visits | 0.25 | 0.00 | 4 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 05's Dominant Sequence
— 1 of 6 instances
1
Select Narrative
2
Acrobat
· reference: Show More Details
3
Microsoft Word
This analyst exhibits 6 distinct sequences across their 6 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 49 | 4.5 | 75.8 | Zero-Edit Visits | 0.19 | 0.00 | 191 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 10's Dominant Sequence
— 8 of 49 instances
1
Select Narrative
2
Show More Details
This analyst exhibits 34 distinct sequences across their 49 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 22 | 2.4 | 72.5 | Zero-Edit Visits | 0.33 | 0.01 | 121 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 13's Dominant Sequence
— 4 of 22 instances
1
Select Narrative
This analyst exhibits 18 distinct sequences across their 22 instances; the dominant one is shown.
|
||||||||
79.8
Avg Adherence Score (0-100)
Median 82.5 · Range 33.1–97.5
2.0 min
Avg Task Duration
0.5 min median
3.6
Avg Sub-Steps per Instance
75.0% have multiple sub-steps
16.5%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
16.5% of instances follow this exactly
Select Narrative Template — Mined SoP
The most common sub-step sequence observed across 600 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Select Narrative
0.6s
0.1 edits
Veeva Safety
100%
2
Show More Details
0.3s
0.1 edits
Veeva Safety
43%
Variant Explorer
Reference
Select Narrative Template Reference Path
99 instances · 9 analysts
This is the mined reference path — 99 instances across 9 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Select Narrative Template.
Genuine Complexity
Complex-Case Execution Path
150 instances · 11 analysts
· Δ adherence +4.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
×
Show More Details
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 84.3 vs reference 79.8 (delta +4.5). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Reviewer Sweep
Reviewer Verification Pass
7 instances · 2 analysts
· Δ adherence +2.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
View Migration Notice
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 2 analysts across 6 cases.
Genuine Complexity
Complex-Case Execution Path
6 instances · 3 analysts
· Δ adherence +12.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Preview Japanese Narrative
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 92.5 vs reference 79.8 (delta +12.7). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
5 instances · 2 analysts
· Δ adherence +4.2
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Select Documents
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 84.0 vs reference 79.8 (delta +4.2). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Word
5 instances · 3 analysts
· Δ adherence -9.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
Microsoft Word
extra step (reference ends at step 2)
Cross-App Swivel — 3 analysts pivot into Microsoft Word during this task (0% of task time in non-primary apps, 40% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Cross-App Swivel
Cross-App Lookup to Microsoft Teams
4 instances · 2 analysts
· Δ adherence -7.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
Microsoft Teams
extra step (reference ends at step 2)
Cross-App Swivel — 2 analysts pivot into Microsoft Teams during this task (0% of task time in non-primary apps, 33% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Genuine Complexity
Complex-Case Execution Path
4 instances · 2 analysts
· Δ adherence +6.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Preview English Narrative
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 86.2 vs reference 79.8 (delta +6.4). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Word
4 instances · 3 analysts
· Δ adherence -12.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Microsoft Word
reference at this step: Show More Details
Cross-App Swivel — 3 analysts pivot into Microsoft Word during this task (0% of task time in non-primary apps, 50% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Cross-App Swivel
Cross-App Lookup to Acrobat
4 instances · 1 analysts
· Δ adherence -7.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
Acrobat
extra step (reference ends at step 2)
Cross-App Swivel — 1 analysts pivot into Acrobat during this task (0% of task time in non-primary apps, 33% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Reviewer Sweep
Reviewer Verification Pass
4 instances · 2 analysts
· Δ adherence +2.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
View Investigator Information
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 2 analysts across 3 cases.
Reviewer Sweep
Reviewer Verification Pass
3 instances · 1 analysts
· Δ adherence +2.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
View Physician Report
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 1 analysts across 3 cases.
Genuine Complexity
Complex-Case Execution Path
3 instances · 3 analysts
· Δ adherence +4.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
View Narrative Preview Header
extra step (reference ends at step 2)
Genuine Complexity — this sequence differs from the reference path (3 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 84.2 vs reference 79.8 (delta +4.3). 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 | 1,702 | 78.7% | 87.1% |
| Microsoft Word | 152 | 7.0% | 0.0% |
| Acrobat | 99 | 4.6% | 0.0% |
| Microsoft Teams | 53 | 2.4% | 0.0% |
| gpteal.merck.com | 31 | 1.4% | 0.0% |
| Notepad | 22 | 1.0% | 0.0% |
| collaboration.merck.com | 19 | 0.9% | 0.0% |
| Microsoft Excel | 17 | 0.8% | 0.0% |
| Microsoft Outlook | 14 | 0.6% | 0.0% |
| CelonisTaskMining.Client | 6 | 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 | 67 |
| Microsoft Word | Acrobat | 57 |
| Acrobat | Microsoft Word | 52 |
| Microsoft Word | Veeva Safety | 42 |
| Veeva Safety | Microsoft Teams | 31 |
| Veeva Safety | Acrobat | 31 |
| Veeva Safety | gpteal.merck.com | 22 |
| Acrobat | Veeva Safety | 20 |
| Veeva Safety | Notepad | 17 |
| Veeva Safety | collaboration.merck.com | 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 7,023 events 318 cases 70.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<li>
"Narrative"
683×
pyzeClick
<span>
"Complete"
593×
pyzeClick
<a>
"Show more"
520×
pyzeClick
<li>
"Details"
288×
pyzeClick
<button>
"Create"
131×
pyzeClick
<li>
"Patient"
122×
pyzeClick
<span>
"Validation Results (Failures & Warnings)"
117×
pyzeClick
<span>
"Medical History & Concurrent Conditions"
115×
Phobos DOM 1,104 events 155 cases 30.0 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
153×
pyzeClick
<button>
"Advance"
89×
pyzeClick
<button>
"Save Comments"
77×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
52×
pyzeClick
<button>
"Send to process"
43×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
40×
pyzeClick
<a>
"Cases"
35×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
34×
Microsoft Word Document 441 events 128 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 540591 (English) (26495494_0_2_345212)
726×
file
V-Narrative 533164 (English) (26378749_0_4_342805)
659×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
572×
file
V-Narrative 553016 (English) (26689302_0_2_348778)
392×
file
V-Narrative 530932 (English) (26349278_0_2_339374)
387×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
382×
file
V-Narrative 533533 (English) (26384397_0_2_337752)
361×
file
V-Narrative 525025 (English) (26253159_0_3_342205)
320×
Acrobat App 289 events 71 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,900 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 192 events 87 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,339 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 100 events 52 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Daily tracking list (PV-SMART) copy
523×
file
E2B_MSDPRO_014172_AGPROD1_20251127163042
155×
file
Priorización
152×
file
I-0000350382
142×
file
records_as_of_2025_11_28_EST
140×
file
MERCK2516915_ARGPRO_20251125054618
111×
file
Case tracking list
82×
file
MERCK2524493_ARGPRO_20251202045619
46×
Microsoft Outlook Message 72 events 44 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
collaboration.merck.com App 68 events 37 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
gpteal.merck.com DOM 56 events 33 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Notepad App 54 events 30 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.
0.11
Swivel Rate
events in non-primary app / total
0.92
Zero-Edit Visit Rate
inspection-only clicks
1.29
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 1.29
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 |
|---|---|---|---|---|---|---|---|
| 2335834 | Analyst 01 | 33.1 | 8.2m | 29 | 0.90 | 0.00 | 13 |
| 2290475 | Analyst 01 | 35.1 | 14.7m | 35 | 0.83 | 0.00 | 14 |
| 2350181 | Analyst 01 | 39.5 | 26.4m | 33 | 0.70 | 0.00 | 11 |
| 2309351 | Analyst 01 | 45.8 | 16.0m | 18 | 0.72 | 0.00 | 7 |
| 2351603 | Analyst 10 | 51.0 | 13.2m | 10 | 0.80 | 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) →
15
11
07
04
14
08
03
12
01
09
02
05
10
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 15 | 5 | 5.0 | 90.3 | Zero-Edit Visits | 0.00 | 0.00 | 12 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 15's Dominant Sequence
— 4 of 5 instances
1
Select Narrative
This analyst exhibits 2 distinct sequences across their 5 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 3 | 3.0 | 87.5 | Zero-Edit Visits | 0.00 | 0.00 | 3 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 11's Dominant Sequence
— 3 of 3 instances
1
Select Narrative
|
||||||||
| ▸ | Analyst 07 | 1 | 1.0 | 85.5 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 07's Dominant Sequence
— 1 of 1 instances
1
Select Narrative
2
Show More Details
3
View Narrative Preview Header
4
Update Case Narrative
5
Select List Item
|
||||||||
| ▸ | Analyst 04 | 6 | 1.2 | 85.0 | Zero-Edit Visits | 0.00 | 0.00 | 19 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 04's Dominant Sequence
— 4 of 6 instances
1
Select Narrative
This analyst exhibits 3 distinct sequences across their 6 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 9 | 4.5 | 83.8 | Zero-Edit Visits | 0.00 | 0.00 | 20 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 14's Dominant Sequence
— 4 of 9 instances
1
Select Narrative
This analyst exhibits 6 distinct sequences across their 9 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 08 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 08's Dominant Sequence
— 1 of 1 instances
1
Select Narrative
2
Show More Details
|
||||||||
| ▸ | Analyst 03 | 20 | 2.0 | 82.0 | Zero-Edit Visits | 0.02 | 0.00 | 11 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 03's Dominant Sequence
— 10 of 20 instances
1
Select Narrative
This analyst exhibits 7 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 30 | 4.3 | 80.5 | Zero-Edit Visits | 0.13 | 0.00 | 101 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 12's Dominant Sequence
— 8 of 30 instances
1
Select Narrative
This analyst exhibits 23 distinct sequences across their 30 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 207 | 13.8 | 80.1 | Zero-Edit Visits | 0.10 | 0.00 | 466 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 01's Dominant Sequence
— 59 of 207 instances
1
Select Narrative
This analyst exhibits 110 distinct sequences across their 207 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 20 | 2.0 | 80.1 | Zero-Edit Visits | 0.07 | 0.00 | 68 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 09's Dominant Sequence
— 6 of 20 instances
1
Select Narrative
This analyst exhibits 11 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 223 | 18.6 | 80.1 | Zero-Edit Visits | 0.09 | 0.00 | 267 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 02's Dominant Sequence
— 49 of 223 instances
1
Select Narrative
2
Show More Details
This analyst exhibits 109 distinct sequences across their 223 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 6 | 1.2 | 77.9 | Zero-Edit Visits | 0.25 | 0.00 | 4 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 05's Dominant Sequence
— 1 of 6 instances
1
Select Narrative
2
Acrobat
· reference: Show More Details
3
Microsoft Word
This analyst exhibits 6 distinct sequences across their 6 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 47 | 4.3 | 75.6 | Zero-Edit Visits | 0.19 | 0.00 | 171 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 10's Dominant Sequence
— 9 of 47 instances
1
Select Narrative
2
Show More Details
This analyst exhibits 32 distinct sequences across their 47 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 22 | 2.4 | 73.6 | Zero-Edit Visits | 0.31 | 0.00 | 65 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 13's Dominant Sequence
— 4 of 22 instances
1
Select Narrative
This analyst exhibits 18 distinct sequences across their 22 instances; the dominant one is shown.
|
||||||||
79.7
Avg Adherence Score (0-100)
Median 82.5 · Range 33.1–97.5
2.5 min
Avg Task Duration
0.5 min median
3.8
Avg Sub-Steps per Instance
75.6% have multiple sub-steps
16.3%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
16.3% of instances follow this exactly
Select Narrative Template — Mined SoP
The most common sub-step sequence observed across 595 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Select Narrative
0.6s
0.1 edits
Veeva Safety
100%
2
Show More Details
0.3s
0.1 edits
Veeva Safety
44%
Variant Explorer
Reference
Select Narrative Template Reference Path
97 instances · 9 analysts
This is the mined reference path — 97 instances across 9 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Select Narrative Template.
Genuine Complexity
Complex-Case Execution Path
145 instances · 11 analysts
· Δ adherence +4.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
×
Show More Details
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 84.4 vs reference 79.7 (delta +4.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Linked Remediation Opportunities
Genuine Complexity
Complex-Case Execution Path
6 instances · 3 analysts
· Δ adherence +12.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Preview Japanese Narrative
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 92.5 vs reference 79.7 (delta +12.8). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Linked Remediation Opportunities
Reviewer Sweep
Reviewer Verification Pass
6 instances · 2 analysts
· Δ adherence +2.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
View Migration Notice
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 2 analysts across 5 cases.
Genuine Complexity
Complex-Case Execution Path
4 instances · 2 analysts
· Δ adherence +4.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Select Documents
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 84.4 vs reference 79.7 (delta +4.7). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Linked Remediation Opportunities
Cross-App Swivel
Cross-App Lookup to Microsoft Teams
4 instances · 2 analysts
· Δ adherence -7.2
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
Microsoft Teams
extra step (reference ends at step 2)
Cross-App Swivel — 2 analysts pivot into Microsoft Teams during this task (0% of task time in non-primary apps, 33% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Genuine Complexity
Complex-Case Execution Path
4 instances · 2 analysts
· Δ adherence +6.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Preview English Narrative
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 86.2 vs reference 79.7 (delta +6.5). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Linked Remediation Opportunities
Cross-App Swivel
Cross-App Lookup to Acrobat
4 instances · 1 analysts
· Δ adherence -7.2
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
Acrobat
extra step (reference ends at step 2)
Cross-App Swivel — 1 analysts pivot into Acrobat during this task (0% of task time in non-primary apps, 33% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Reviewer Sweep
Reviewer Verification Pass
4 instances · 2 analysts
· Δ adherence +2.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
View Investigator Information
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 2 analysts across 3 cases.
Genuine Complexity
Complex-Case Execution Path
3 instances · 3 analysts
· Δ adherence +4.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
View Narrative Preview Header
extra step (reference ends at step 2)
Genuine Complexity — this sequence differs from the reference path (3 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 84.2 vs reference 79.7 (delta +4.4). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Linked Remediation Opportunities
Reviewer Sweep
Reviewer Verification Pass
3 instances · 2 analysts
· Δ adherence +0.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
View Validation Results
reference at this step: Show More Details
3
Select Narrative
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 2 analysts across 3 cases.
Reviewer Sweep
Reviewer Verification Pass
3 instances · 2 analysts
· Δ adherence +2.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Show More Details
3
Select Adverse Events
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 2 analysts across 3 cases.
Genuine Complexity
Complex-Case Execution Path
3 instances · 2 analysts
· Δ adherence +2.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Narrative
2
Open Case Assessment Results
reference at this step: Show More Details
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 82.5 vs reference 79.7 (delta +2.8). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Linked Remediation Opportunities
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 | 1,776 | 78.4% | 87.5% |
| Microsoft Word | 165 | 7.3% | 0.0% |
| Acrobat | 101 | 4.5% | 0.0% |
| Microsoft Teams | 56 | 2.5% | 0.0% |
| gpteal.merck.com | 33 | 1.5% | 0.0% |
| Notepad | 22 | 1.0% | 0.0% |
| collaboration.merck.com | 21 | 0.9% | 0.0% |
| Microsoft Excel | 19 | 0.8% | 0.0% |
| Microsoft Outlook | 14 | 0.6% | 0.0% |
| dtgpteal.merck.com | 6 | 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 | 72 |
| Microsoft Word | Acrobat | 59 |
| Microsoft Word | Veeva Safety | 52 |
| Acrobat | Microsoft Word | 52 |
| Veeva Safety | Microsoft Teams | 33 |
| Veeva Safety | Acrobat | 31 |
| Acrobat | Veeva Safety | 24 |
| Veeva Safety | gpteal.merck.com | 22 |
| Veeva Safety | Notepad | 17 |
| Microsoft Teams | Veeva Safety | 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 7,023 events 318 cases 70.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<li>
"Narrative"
683×
pyzeClick
<span>
"Complete"
593×
pyzeClick
<a>
"Show more"
520×
pyzeClick
<li>
"Details"
288×
pyzeClick
<button>
"Create"
131×
pyzeClick
<li>
"Patient"
122×
pyzeClick
<span>
"Validation Results (Failures & Warnings)"
117×
pyzeClick
<button>
"Save"
115×
Phobos DOM 1,104 events 155 cases 30.0 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
153×
pyzeClick
<button>
"Advance"
89×
pyzeClick
<button>
"Save Comments"
77×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
52×
pyzeClick
<button>
"Send to process"
43×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
40×
pyzeClick
<a>
"Cases"
35×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
34×
Microsoft Word Document 441 events 128 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 540591 (English) (26495494_0_2_345212)
726×
file
V-Narrative 533164 (English) (26378749_0_4_342805)
659×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
572×
file
V-Narrative 553016 (English) (26689302_0_2_348778)
392×
file
V-Narrative 530932 (English) (26349278_0_2_339374)
387×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
382×
file
V-Narrative 533533 (English) (26384397_0_2_337752)
361×
file
V-Narrative 525025 (English) (26253159_0_3_342205)
320×
Acrobat App 289 events 71 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,900 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 192 events 87 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,339 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 100 events 52 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Daily tracking list (PV-SMART) copy
523×
file
E2B_MSDPRO_014172_AGPROD1_20251127163042
155×
file
Priorización
152×
file
I-0000350382
142×
file
records_as_of_2025_11_28_EST
140×
file
MERCK2516915_ARGPRO_20251125054618
111×
file
Case tracking list
82×
file
MERCK2524493_ARGPRO_20251202045619
46×
Microsoft Outlook Message 72 events 44 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
collaboration.merck.com App 68 events 37 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
gpteal.merck.com DOM 56 events 33 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Notepad App 54 events 30 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.
0.11
Swivel Rate
events in non-primary app / total
0.92
Zero-Edit Visit Rate
inspection-only clicks
1.33
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 1.33
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 |
|---|---|---|---|---|---|---|---|
| 2335834 | Analyst 01 | 33.1 | 8.2m | 29 | 0.90 | 0.00 | 13 |
| 2290475 | Analyst 01 | 35.1 | 14.7m | 35 | 0.83 | 0.00 | 14 |
| 2350181 | Analyst 01 | 39.5 | 26.4m | 33 | 0.70 | 0.00 | 11 |
| 2309351 | Analyst 01 | 46.6 | 17.6m | 22 | 0.64 | 0.00 | 8 |
| 2351603 | Analyst 10 | 51.0 | 13.2m | 10 | 0.80 | 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) →
15
11
07
04
14
08
03
02
09
12
01
05
10
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 15 | 5 | 5.0 | 90.3 | Zero-Edit Visits | 0.00 | 0.00 | 12 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 15's Dominant Sequence
— 4 of 5 instances
1
Select Narrative
This analyst exhibits 2 distinct sequences across their 5 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 3 | 3.0 | 87.5 | Zero-Edit Visits | 0.00 | 0.00 | 3 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 11's Dominant Sequence
— 3 of 3 instances
1
Select Narrative
|
||||||||
| ▸ | Analyst 07 | 1 | 1.0 | 85.5 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 07's Dominant Sequence
— 1 of 1 instances
1
Select Narrative
2
Show More Details
3
View Narrative Preview Header
4
Update Case Narrative
5
Select List Item
|
||||||||
| ▸ | Analyst 04 | 6 | 1.2 | 85.0 | Zero-Edit Visits | 0.00 | 0.00 | 19 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 04's Dominant Sequence
— 4 of 6 instances
1
Select Narrative
This analyst exhibits 3 distinct sequences across their 6 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 9 | 4.5 | 83.8 | Zero-Edit Visits | 0.00 | 0.00 | 29 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 14's Dominant Sequence
— 3 of 9 instances
1
Select Narrative
This analyst exhibits 7 distinct sequences across their 9 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 08 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 08's Dominant Sequence
— 1 of 1 instances
1
Select Narrative
2
Show More Details
|
||||||||
| ▸ | Analyst 03 | 20 | 2.0 | 82.0 | Zero-Edit Visits | 0.02 | 0.00 | 11 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 03's Dominant Sequence
— 10 of 20 instances
1
Select Narrative
This analyst exhibits 7 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 222 | 18.5 | 80.1 | Zero-Edit Visits | 0.09 | 0.00 | 329 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 02's Dominant Sequence
— 49 of 222 instances
1
Select Narrative
2
Show More Details
This analyst exhibits 112 distinct sequences across their 222 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 20 | 2.0 | 80.1 | Zero-Edit Visits | 0.07 | 0.00 | 68 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 09's Dominant Sequence
— 6 of 20 instances
1
Select Narrative
This analyst exhibits 11 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 30 | 4.3 | 80.0 | Zero-Edit Visits | 0.14 | 0.00 | 132 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 12's Dominant Sequence
— 7 of 30 instances
1
Select Narrative
This analyst exhibits 24 distinct sequences across their 30 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 203 | 13.5 | 80.0 | Zero-Edit Visits | 0.10 | 0.00 | 545 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 01's Dominant Sequence
— 56 of 203 instances
1
Select Narrative
This analyst exhibits 113 distinct sequences across their 203 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 6 | 1.2 | 77.9 | Zero-Edit Visits | 0.25 | 0.00 | 4 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 05's Dominant Sequence
— 1 of 6 instances
1
Select Narrative
2
Acrobat
· reference: Show More Details
3
Microsoft Word
This analyst exhibits 6 distinct sequences across their 6 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 10 | 47 | 4.3 | 75.5 | Zero-Edit Visits | 0.20 | 0.00 | 191 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 10's Dominant Sequence
— 8 of 47 instances
1
Select Narrative
2
Show More Details
This analyst exhibits 34 distinct sequences across their 47 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 22 | 2.4 | 72.5 | Zero-Edit Visits | 0.33 | 0.01 | 121 |
|
Reference SoP
1
Select Narrative
2
Show More Details
Analyst 13's Dominant Sequence
— 4 of 22 instances
1
Select Narrative
This analyst exhibits 18 distinct sequences across their 22 instances; the dominant one is shown.
|
||||||||