Task-Boundary Mode
How task-instance boundaries are drawn from the event stream. Applies to every Task SoP, Step SoP, and Variants view.
Case Review
📝 Task SoP
Mined from 269 instances
Review SUSAR Case
How 8 analysts execute this task across 91 cases — auto-generated Standard Operating Procedure with adherence scoring.
80.1
Avg Adherence Score (0-100)
Median 82.5 · Range 31.9–97.5
7.6 min
Avg Task Duration
0.6 min median
3.0
Avg Sub-Steps per Instance
68.4% have multiple sub-steps
31.6%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
31.6% of instances follow this exactly
Review SUSAR Case — Mined SoP
The most common sub-step sequence observed across 376 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
View SUSAR Case
0.7s
0.3 edits
Veeva Safety
100%
Variant Explorer
Reference
Review SUSAR Case Reference Path
119 instances · 8 analysts
This is the mined reference path — 119 instances across 8 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Review SUSAR Case.
Genuine Complexity
Complex-Case Execution Path
13 instances · 3 analysts
· Δ adherence +10.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select List 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 90.6 vs reference 80.1 (delta +10.5). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
9 instances · 3 analysts
· Δ adherence +9.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select Rechallenge Not Applicable
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 80.1 (delta +9.9). 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 +8.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
View Yes Value
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 88.9 vs reference 80.1 (delta +8.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
7 instances · 1 analysts
· Δ adherence +9.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select No
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 80.1 (delta +9.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Acrobat
6 instances · 4 analysts
· Δ adherence -12.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Acrobat
extra step (reference ends at step 1)
Cross-App Swivel — 4 analysts pivot into Acrobat 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 Microsoft Excel
5 instances · 2 analysts
· Δ adherence -12.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Microsoft Excel
extra step (reference ends at step 1)
Cross-App Swivel — 2 analysts pivot into Microsoft Excel 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.
Genuine Complexity
Complex-Case Execution Path
5 instances · 1 analysts
· Δ adherence +8.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
View No Value
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 88.5 vs reference 80.1 (delta +8.4). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
4 instances · 1 analysts
· Δ adherence +2.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Generate Final Review Form
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 80.1 (delta +2.4). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
4 instances · 1 analysts
· Δ adherence +9.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select Unknown Value
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 80.1 (delta +9.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Phobos
3 instances · 2 analysts
· Δ adherence -17.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Add New Case
extra step (reference ends at step 1)
Cross-App Swivel — 2 analysts pivot into Phobos during this task (0% of task time in non-primary apps, 61% 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
3 instances · 1 analysts
· Δ adherence +7.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select Linked Cases
extra step (reference ends at step 1)
3
Select List Item
extra step (reference ends at step 1)
Genuine Complexity — this sequence differs from the reference path (3 sub-steps vs 1 in reference) but shows no strong pain-point signals: adherence score 87.5 vs reference 80.1 (delta +7.4). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Word
3 instances · 3 analysts
· Δ adherence -10.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Microsoft Word
extra step (reference ends at step 1)
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.
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 | 834 | 72.8% | 95.5% |
| Acrobat | 57 | 5.0% | 0.0% |
| collaboration.merck.com | 52 | 4.5% | 0.0% |
| usc-excel.officeapps.live.com | 37 | 3.2% | 0.0% |
| Microsoft Teams | 35 | 3.1% | 0.0% |
| Microsoft Excel | 35 | 3.1% | 0.0% |
| Microsoft Word | 34 | 3.0% | 0.0% |
| YoudaoDict | 13 | 1.1% | 0.0% |
| MEDDRABROWSERWIN | 7 | 0.6% | 0.0% |
| Phobos | 6 | 0.5% | 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 |
|---|---|---|
| collaboration.merck.com | usc-excel.officeapps.live.com | 28 |
| usc-excel.officeapps.live.com | collaboration.merck.com | 26 |
| Veeva Safety | Microsoft Excel | 24 |
| Veeva Safety | Acrobat | 21 |
| Veeva Safety | Microsoft Teams | 19 |
| Acrobat | Microsoft Word | 19 |
| Microsoft Word | Acrobat | 18 |
| Acrobat | Veeva Safety | 17 |
| Veeva Safety | collaboration.merck.com | 13 |
| Veeva Safety | Microsoft Word | 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 2,972 events 91 cases 47.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
179×
pyzeClick
<span>
"Narrative"
123×
pyzeClick
<span>
"Adverse Events"
105×
pyzeClick
<a>
"Show more"
91×
pyzeClick
<span>
"Submissions & Distributions"
72×
pyzeClick
<button>
"Create"
66×
pyzeClick
<li>
"Narrative"
64×
pyzeClick
<span>
"Reference Numbers"
60×
Phobos DOM 526 events 49 cases 12.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
65×
pyzeClick
<button>
"Save Comments"
56×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
42×
pyzeClick
<button>
"Send to process"
33×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
32×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
30×
pyzeClick
<a>
"Cases"
23×
pyzeClick
<button>
"Advance"
23×
Microsoft Word Document 140 events 38 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 533164 (English) (26378749_0_4_342805)
659×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
515×
file
V-Narrative 543823 (English) (26544362_0_2_343646)
514×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
429×
file
V-Narrative 553016 (English) (26689302_0_2_348778)
392×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
377×
file
V-Narrative 530932 (English) (26349278_0_2_339374)
366×
file
V-Narrative 555782 (English) (26733642_0_2_350941)
205×
Acrobat App 133 events 31 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,046 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.
collaboration.merck.com App 114 events 23 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
363 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 106 events 38 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,233 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 93 events 35 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
usc-excel.officeapps.live.com Document 70 events 10 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Outlook Message 31 events 18 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
apps.powerapps.com App 23 events 4 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.
0.13
Swivel Rate
events in non-primary app / total
0.86
Zero-Edit Visit Rate
inspection-only clicks
1.27
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 0.86
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 0.86
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 1.27
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 |
|---|---|---|---|---|---|---|---|
| 2349107 | Analyst 13 | 31.9 | 5.3m | 32 | 0.94 | 0.00 | 14 |
| 2241891 | Analyst 01 | 32.6 | 19.3m | 41 | 0.93 | 0.00 | 19 |
| 2258699 | Analyst 13 | 32.7 | 0.9m | 22 | 0.91 | 0.00 | 10 |
| 2334048 | Analyst 12 | 43.7 | 1.4m | 17 | 0.82 | 0.00 | 7 |
| 2252232 | Analyst 13 | 44.8 | 6.6m | 13 | 0.92 | 0.00 | 5 |
Users on This Task
8 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
11
09
01
02
05
12
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 04 | 60 | 8.6 | 83.3 | Zero-Edit Visits | 0.00 | 0.00 | 740 |
|
Reference SoP
1
View SUSAR Case
Analyst 04's Dominant Sequence
— 27 of 60 instances
1
View SUSAR Case
This analyst exhibits 28 distinct sequences across their 60 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 3 | 3.0 | 83.1 | Zero-Edit Visits | 0.00 | 0.00 | 226 |
|
Reference SoP
1
View SUSAR Case
Analyst 11's Dominant Sequence
— 1 of 3 instances
1
View SUSAR Case
This analyst exhibits 3 distinct sequences across their 3 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 147 | 12.2 | 82.2 | Zero-Edit Visits | 0.12 | 0.00 | 346 |
|
Reference SoP
1
View SUSAR Case
Analyst 09's Dominant Sequence
— 36 of 147 instances
1
View SUSAR Case
This analyst exhibits 72 distinct sequences across their 147 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 25 | 3.6 | 80.3 | Zero-Edit Visits | 0.11 | 0.00 | 71 |
|
Reference SoP
1
View SUSAR Case
Analyst 01's Dominant Sequence
— 6 of 25 instances
1
View SUSAR Case
This analyst exhibits 18 distinct sequences across their 25 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 64 | 8.0 | 79.6 | Zero-Edit Visits | 0.12 | 0.01 | 80 |
|
Reference SoP
1
View SUSAR Case
Analyst 02's Dominant Sequence
— 25 of 64 instances
1
View SUSAR Case
This analyst exhibits 36 distinct sequences across their 64 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 9 | 1.8 | 78.4 | Zero-Edit Visits | 0.21 | 0.00 | 10 |
|
Reference SoP
1
View SUSAR Case
Analyst 05's Dominant Sequence
— 3 of 9 instances
1
View SUSAR Case
This analyst exhibits 7 distinct sequences across their 9 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 25 | 6.2 | 75.1 | Zero-Edit Visits | 0.23 | 0.00 | 100 |
|
Reference SoP
1
View SUSAR Case
Analyst 12's Dominant Sequence
— 9 of 25 instances
1
View SUSAR Case
This analyst exhibits 16 distinct sequences across their 25 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 43 | 4.8 | 71.7 | Zero-Edit Visits | 0.32 | 0.00 | 1,300 |
|
Reference SoP
1
View SUSAR Case
Analyst 13's Dominant Sequence
— 12 of 43 instances
1
View SUSAR Case
This analyst exhibits 30 distinct sequences across their 43 instances; the dominant one is shown.
|
||||||||
80.0
Avg Adherence Score (0-100)
Median 82.5 · Range 31.1–97.5
12.7 min
Avg Task Duration
0.8 min median
3.4
Avg Sub-Steps per Instance
72.1% have multiple sub-steps
0.5%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
0.5% of instances follow this exactly
Review SUSAR Case — Mined SoP
The most common sub-step sequence observed across 376 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
View SUSAR Case
0.7s
0.3 edits
Veeva Safety
100%
2
Open Study Section
0.0s
Veeva Safety
1%
Variant Explorer
Reference
Review SUSAR Case Reference Path
2 instances · 1 analysts
This is the mined reference path — 2 instances across 1 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Review SUSAR Case.
Genuine Complexity
Complex-Case Execution Path
105 instances · 8 analysts
· Δ adherence +2.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
×
Open Study Section
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 82.9 vs reference 80.0 (delta +2.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
13 instances · 3 analysts
· Δ adherence +10.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select List Item
reference at this step: Open Study Section
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 90.6 vs reference 80.0 (delta +10.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
9 instances · 3 analysts
· Δ adherence +10.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select Rechallenge Not Applicable
reference at this step: Open Study Section
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 90.0 vs reference 80.0 (delta +10.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
7 instances · 1 analysts
· Δ adherence +10.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select No
reference at this step: Open Study Section
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 90.0 vs reference 80.0 (delta +10.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
6 instances · 2 analysts
· Δ adherence +10.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
View Yes Value
reference at this step: Open Study Section
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 90.0 vs reference 80.0 (delta +10.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Acrobat
5 instances · 4 analysts
· Δ adherence -12.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Acrobat
reference at this step: Open Study Section
Cross-App Swivel — 4 analysts pivot into Acrobat 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.
Genuine Complexity
Complex-Case Execution Path
5 instances · 1 analysts
· Δ adherence +8.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
View No Value
reference at this step: Open Study Section
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 88.5 vs reference 80.0 (delta +8.5). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Excel
4 instances · 2 analysts
· Δ adherence -12.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Microsoft Excel
reference at this step: Open Study Section
Cross-App Swivel — 2 analysts pivot into Microsoft Excel 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.
Genuine Complexity
Complex-Case Execution Path
4 instances · 1 analysts
· Δ adherence +2.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Generate Final Review Form
reference at this step: Open Study Section
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
4 instances · 2 analysts
· Δ adherence +4.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Delete Item
reference at this step: Open Study Section
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.
Genuine Complexity
Complex-Case Execution Path
4 instances · 1 analysts
· Δ adherence +10.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select Unknown Value
reference at this step: Open Study Section
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 90.0 vs reference 80.0 (delta +10.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Phobos
3 instances · 2 analysts
· Δ adherence -17.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Add New Case
reference at this step: Open Study Section
Cross-App Swivel — 2 analysts pivot into Phobos during this task (0% of task time in non-primary apps, 61% 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.
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 | 911 | 72.1% | 96.1% |
| collaboration.merck.com | 64 | 5.1% | 0.0% |
| Acrobat | 62 | 4.9% | 0.0% |
| usc-excel.officeapps.live.com | 49 | 3.9% | 0.0% |
| Microsoft Excel | 40 | 3.2% | 0.0% |
| Microsoft Word | 37 | 2.9% | 0.0% |
| Microsoft Teams | 36 | 2.8% | 0.0% |
| YoudaoDict | 13 | 1.0% | 0.0% |
| MEDDRABROWSERWIN | 8 | 0.6% | 0.0% |
| Phobos | 7 | 0.6% | 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 |
|---|---|---|
| collaboration.merck.com | usc-excel.officeapps.live.com | 39 |
| usc-excel.officeapps.live.com | collaboration.merck.com | 34 |
| Veeva Safety | Microsoft Excel | 28 |
| Veeva Safety | Acrobat | 22 |
| Acrobat | Microsoft Word | 21 |
| Acrobat | Veeva Safety | 20 |
| Microsoft Word | Acrobat | 19 |
| Veeva Safety | Microsoft Teams | 19 |
| Microsoft Excel | Veeva Safety | 14 |
| Veeva Safety | collaboration.merck.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 2,972 events 91 cases 47.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
179×
pyzeClick
<span>
"Narrative"
123×
pyzeClick
<span>
"Adverse Events"
105×
pyzeClick
<a>
"Show more"
91×
pyzeClick
<span>
"Submissions & Distributions"
72×
pyzeClick
<button>
"Create"
66×
pyzeClick
<li>
"Narrative"
64×
pyzeClick
<span>
"Reference Numbers"
60×
Phobos DOM 526 events 49 cases 12.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
65×
pyzeClick
<button>
"Save Comments"
56×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
42×
pyzeClick
<button>
"Send to process"
33×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
32×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
30×
pyzeClick
<a>
"Cases"
23×
pyzeClick
<button>
"Advance"
23×
Microsoft Word Document 140 events 38 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 533164 (English) (26378749_0_4_342805)
659×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
515×
file
V-Narrative 543823 (English) (26544362_0_2_343646)
514×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
429×
file
V-Narrative 553016 (English) (26689302_0_2_348778)
392×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
377×
file
V-Narrative 530932 (English) (26349278_0_2_339374)
366×
file
V-Narrative 555782 (English) (26733642_0_2_350941)
205×
Acrobat App 133 events 31 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,046 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.
collaboration.merck.com App 114 events 23 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
363 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 106 events 38 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,233 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 93 events 35 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
usc-excel.officeapps.live.com Document 70 events 10 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Outlook Message 31 events 18 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
apps.powerapps.com App 23 events 4 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.
0.13
Swivel Rate
events in non-primary app / total
0.85
Zero-Edit Visit Rate
inspection-only clicks
1.31
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 0.85
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 0.85
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 1.31
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 |
|---|---|---|---|---|---|---|---|
| 2335072 | Analyst 13 | 31.1 | 7.2m | 27 | 0.96 | 0.00 | 12 |
| 2349107 | Analyst 13 | 31.9 | 5.3m | 32 | 0.94 | 0.00 | 14 |
| 2258699 | Analyst 13 | 32.7 | 0.9m | 22 | 0.91 | 0.00 | 10 |
| 2241891 | Analyst 01 | 34.0 | 20.6m | 45 | 0.89 | 0.00 | 20 |
| 2334048 | Analyst 12 | 43.7 | 1.4m | 17 | 0.82 | 0.00 | 7 |
Users on This Task
8 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
11
09
01
02
05
12
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 04 | 60 | 8.6 | 83.3 | Zero-Edit Visits | 0.00 | 0.00 | 789 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 04's Dominant Sequence
— 24 of 60 instances
1
View SUSAR Case
This analyst exhibits 29 distinct sequences across their 60 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 3 | 3.0 | 83.1 | Zero-Edit Visits | 0.00 | 0.00 | 226 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 11's Dominant Sequence
— 1 of 3 instances
1
View SUSAR Case
This analyst exhibits 3 distinct sequences across their 3 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 147 | 12.2 | 82.1 | Zero-Edit Visits | 0.12 | 0.00 | 2,064 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 09's Dominant Sequence
— 31 of 147 instances
1
View SUSAR Case
This analyst exhibits 77 distinct sequences across their 147 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 25 | 3.6 | 80.5 | Zero-Edit Visits | 0.11 | 0.00 | 83 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 01's Dominant Sequence
— 4 of 25 instances
1
View SUSAR Case
This analyst exhibits 20 distinct sequences across their 25 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 64 | 8.0 | 79.6 | Zero-Edit Visits | 0.12 | 0.01 | 94 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 02's Dominant Sequence
— 24 of 64 instances
1
View SUSAR Case
This analyst exhibits 37 distinct sequences across their 64 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 9 | 1.8 | 78.4 | Zero-Edit Visits | 0.21 | 0.00 | 10 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 05's Dominant Sequence
— 3 of 9 instances
1
View SUSAR Case
This analyst exhibits 7 distinct sequences across their 9 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 25 | 6.2 | 75.6 | Zero-Edit Visits | 0.23 | 0.00 | 128 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 12's Dominant Sequence
— 8 of 25 instances
1
View SUSAR Case
This analyst exhibits 17 distinct sequences across their 25 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 43 | 4.8 | 71.2 | Zero-Edit Visits | 0.33 | 0.00 | 1,362 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 13's Dominant Sequence
— 10 of 43 instances
1
View SUSAR Case
This analyst exhibits 33 distinct sequences across their 43 instances; the dominant one is shown.
|
||||||||
77.1
Avg Adherence Score (0-100)
Median 82.5 · Range 31.9–97.5
10.3 min
Avg Task Duration
1.3 min median
4.1
Avg Sub-Steps per Instance
64.5% have multiple sub-steps
35.5%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
35.5% of instances follow this exactly
Review SUSAR Case — Mined SoP
The most common sub-step sequence observed across 279 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
View SUSAR Case
0.7s
0.3 edits
Veeva Safety
100%
Variant Explorer
Reference
Review SUSAR Case Reference Path
99 instances · 8 analysts
This is the mined reference path — 99 instances across 8 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Review SUSAR Case.
Cross-App Swivel
Cross-App Lookup to Acrobat
5 instances · 3 analysts
· Δ adherence -9.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Acrobat
extra step (reference ends at step 1)
Cross-App Swivel — 3 analysts pivot into Acrobat 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 Microsoft Excel
4 instances · 2 analysts
· Δ adherence -9.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Microsoft Excel
extra step (reference ends at step 1)
Cross-App Swivel — 2 analysts pivot into Microsoft Excel 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 Phobos
3 instances · 2 analysts
· Δ adherence -14.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Add New Case
extra step (reference ends at step 1)
Cross-App Swivel — 2 analysts pivot into Phobos during this task (0% of task time in non-primary apps, 61% 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
3 instances · 1 analysts
· Δ adherence +6.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select Linked Cases
extra step (reference ends at step 1)
3
Select List Item
extra step (reference ends at step 1)
4
View SUSAR Case
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 83.8 vs reference 77.1 (delta +6.7). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Word
3 instances · 3 analysts
· Δ adherence -7.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Microsoft Word
extra step (reference ends at step 1)
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.
Genuine Complexity
Complex-Case Execution Path
3 instances · 1 analysts
· Δ adherence +5.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Generate Final Review Form
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 77.1 (delta +5.4). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to collaboration.merck.com
3 instances · 1 analysts
· Δ adherence -9.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
collaboration.merck.com
extra step (reference ends at step 1)
Cross-App Swivel — 1 analysts pivot into collaboration.merck.com 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.
Genuine Complexity
Complex-Case Execution Path
3 instances · 2 analysts
· Δ adherence +7.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
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 85.0 vs reference 77.1 (delta +7.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
2 instances · 2 analysts
· Δ adherence +5.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
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 82.5 vs reference 77.1 (delta +5.4). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Amendment Spiral
Post-Amendment Rework
2 instances · 2 analysts
· Δ adherence +7.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select List Item
extra step (reference ends at step 1)
3
View SUSAR Case
extra step (reference ends at step 1)
Amendment Spiral — multiple analysts (2) revisit the same activities on the same cases (2), 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
2 instances · 2 analysts
· Δ adherence +5.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select Start Case Entry
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 77.1 (delta +5.4). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Teams
2 instances · 2 analysts
· Δ adherence -9.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Microsoft Teams
extra step (reference ends at step 1)
Cross-App Swivel — 2 analysts pivot into Microsoft Teams 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.
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 | 834 | 72.8% | 95.5% |
| Acrobat | 57 | 5.0% | 0.0% |
| collaboration.merck.com | 52 | 4.5% | 0.0% |
| usc-excel.officeapps.live.com | 37 | 3.2% | 0.0% |
| Microsoft Teams | 35 | 3.1% | 0.0% |
| Microsoft Excel | 35 | 3.1% | 0.0% |
| Microsoft Word | 34 | 3.0% | 0.0% |
| YoudaoDict | 13 | 1.1% | 0.0% |
| MEDDRABROWSERWIN | 7 | 0.6% | 0.0% |
| Microsoft Outlook | 6 | 0.5% | 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 |
|---|---|---|
| collaboration.merck.com | usc-excel.officeapps.live.com | 28 |
| usc-excel.officeapps.live.com | collaboration.merck.com | 26 |
| Veeva Safety | Microsoft Excel | 24 |
| Veeva Safety | Acrobat | 21 |
| Veeva Safety | Microsoft Teams | 19 |
| Acrobat | Microsoft Word | 19 |
| Microsoft Word | Acrobat | 18 |
| Acrobat | Veeva Safety | 17 |
| Veeva Safety | collaboration.merck.com | 14 |
| Veeva Safety | Microsoft Word | 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 2,972 events 91 cases 47.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
179×
pyzeClick
<span>
"Narrative"
123×
pyzeClick
<span>
"Adverse Events"
105×
pyzeClick
<a>
"Show more"
91×
pyzeClick
<span>
"Submissions & Distributions"
72×
pyzeClick
<button>
"Create"
66×
pyzeClick
<li>
"Narrative"
64×
pyzeClick
<span>
"Reference Numbers"
60×
Phobos DOM 526 events 49 cases 12.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
65×
pyzeClick
<button>
"Save Comments"
56×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
42×
pyzeClick
<button>
"Send to process"
33×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
32×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
30×
pyzeClick
<button>
"Advance"
23×
pyzeClick
<a>
"Cases"
23×
Microsoft Word Document 140 events 38 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 533164 (English) (26378749_0_4_342805)
659×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
515×
file
V-Narrative 543823 (English) (26544362_0_2_343646)
514×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
429×
file
V-Narrative 553016 (English) (26689302_0_2_348778)
392×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
377×
file
V-Narrative 530932 (English) (26349278_0_2_339374)
366×
file
V-Narrative 555782 (English) (26733642_0_2_350941)
205×
Acrobat App 133 events 31 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,046 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.
collaboration.merck.com App 114 events 23 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
363 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 106 events 38 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,233 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 93 events 35 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
usc-excel.officeapps.live.com Document 70 events 10 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Outlook Message 31 events 18 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
apps.powerapps.com App 23 events 4 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.
0.16
Swivel Rate
events in non-primary app / total
0.92
Zero-Edit Visit Rate
inspection-only clicks
1.68
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 0.16
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Revisit Depth
value 1.68
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 |
|---|---|---|---|---|---|---|---|
| 2349107 | Analyst 13 | 31.9 | 5.3m | 32 | 0.94 | 0.00 | 14 |
| 2241891 | Analyst 01 | 32.6 | 19.3m | 41 | 0.93 | 0.00 | 19 |
| 2258699 | Analyst 13 | 32.7 | 0.9m | 22 | 0.91 | 0.00 | 10 |
| 2334048 | Analyst 12 | 36.6 | 1.9m | 25 | 0.80 | 0.00 | 10 |
| 2352746 | Analyst 02 | 40.7 | 7.8m | 21 | 0.95 | 0.24 | 4 |
Users on This Task
8 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) →
11
04
02
05
09
01
12
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 11 | 3 | 3.0 | 83.1 | Zero-Edit Visits | 0.00 | 0.00 | 226 |
|
Reference SoP
1
View SUSAR Case
Analyst 11's Dominant Sequence
— 1 of 3 instances
1
View SUSAR Case
This analyst exhibits 3 distinct sequences across their 3 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 50 | 7.1 | 82.4 | Zero-Edit Visits | 0.00 | 0.00 | 740 |
|
Reference SoP
1
View SUSAR Case
Analyst 04's Dominant Sequence
— 23 of 50 instances
1
View SUSAR Case
This analyst exhibits 24 distinct sequences across their 50 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 53 | 6.6 | 78.5 | Zero-Edit Visits | 0.13 | 0.01 | 82 |
|
Reference SoP
1
View SUSAR Case
Analyst 02's Dominant Sequence
— 21 of 53 instances
1
View SUSAR Case
This analyst exhibits 31 distinct sequences across their 53 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 8 | 1.6 | 77.1 | Zero-Edit Visits | 0.21 | 0.00 | 10 |
|
Reference SoP
1
View SUSAR Case
Analyst 05's Dominant Sequence
— 3 of 8 instances
1
View SUSAR Case
This analyst exhibits 6 distinct sequences across their 8 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 86 | 7.2 | 76.7 | Zero-Edit Visits | 0.19 | 0.00 | 347 |
|
Reference SoP
1
View SUSAR Case
Analyst 09's Dominant Sequence
— 26 of 86 instances
1
View SUSAR Case
This analyst exhibits 56 distinct sequences across their 86 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 17 | 2.4 | 76.7 | Zero-Edit Visits | 0.16 | 0.00 | 71 |
|
Reference SoP
1
View SUSAR Case
Analyst 01's Dominant Sequence
— 5 of 17 instances
1
View SUSAR Case
This analyst exhibits 12 distinct sequences across their 17 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 22 | 5.5 | 74.4 | Zero-Edit Visits | 0.23 | 0.00 | 101 |
|
Reference SoP
1
View SUSAR Case
Analyst 12's Dominant Sequence
— 8 of 22 instances
1
View SUSAR Case
This analyst exhibits 14 distinct sequences across their 22 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 40 | 4.4 | 70.7 | Zero-Edit Visits | 0.33 | 0.00 | 1,300 |
|
Reference SoP
1
View SUSAR Case
Analyst 13's Dominant Sequence
— 12 of 40 instances
1
View SUSAR Case
This analyst exhibits 27 distinct sequences across their 40 instances; the dominant one is shown.
|
||||||||
76.8
Avg Adherence Score (0-100)
Median 82.5 · Range 31.1–97.5
17.7 min
Avg Task Duration
1.7 min median
4.7
Avg Sub-Steps per Instance
68.0% have multiple sub-steps
0.4%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
0.4% of instances follow this exactly
Review SUSAR Case — Mined SoP
The most common sub-step sequence observed across 269 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
View SUSAR Case
0.5s
0.2 edits
Veeva Safety
100%
2
Open Study Section
0.0s
Veeva Safety
0%
Variant Explorer
Reference
Review SUSAR Case Reference Path
1 instances · 1 analysts
This is the mined reference path — 1 instances across 1 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Review SUSAR Case.
Genuine Complexity
Complex-Case Execution Path
86 instances · 8 analysts
· Δ adherence +5.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
×
Open Study Section
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 82.7 vs reference 76.8 (delta +5.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Acrobat
4 instances · 3 analysts
· Δ adherence -9.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Acrobat
reference at this step: Open Study Section
Cross-App Swivel — 3 analysts pivot into Acrobat 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 Microsoft Excel
4 instances · 2 analysts
· Δ adherence -9.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Microsoft Excel
reference at this step: Open Study Section
Cross-App Swivel — 2 analysts pivot into Microsoft Excel 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.
Linked Remediation Opportunities
Genuine Complexity
Complex-Case Execution Path
4 instances · 2 analysts
· Δ adherence +7.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Delete Item
reference at this step: Open Study Section
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 76.8 (delta +7.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Phobos
3 instances · 2 analysts
· Δ adherence -14.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Add New Case
reference at this step: Open Study Section
Cross-App Swivel — 2 analysts pivot into Phobos during this task (0% of task time in non-primary apps, 61% 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
3 instances · 1 analysts
· Δ adherence +7.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select Linked Cases
reference at this step: Open Study Section
3
Select List Item
extra step (reference ends at step 2)
4
View SUSAR Case
extra step (reference ends at step 2)
Genuine Complexity — this sequence differs from the reference path (4 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 83.8 vs reference 76.8 (delta +7.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Word
3 instances · 3 analysts
· Δ adherence -6.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Microsoft Word
reference at this step: Open Study Section
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 collaboration.merck.com
3 instances · 1 analysts
· Δ adherence -9.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
collaboration.merck.com
reference at this step: Open Study Section
Cross-App Swivel — 1 analysts pivot into collaboration.merck.com 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.
Linked Remediation Opportunities
Genuine Complexity
Complex-Case Execution Path
2 instances · 2 analysts
· Δ adherence +5.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Start Approval Workflow
reference at this step: Open Study Section
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 76.8 (delta +5.7). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Amendment Spiral
Post-Amendment Rework
2 instances · 2 analysts
· Δ adherence +8.2
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select List Item
reference at this step: Open Study Section
3
View SUSAR Case
extra step (reference ends at step 2)
Amendment Spiral — multiple analysts (2) revisit the same activities on the same cases (2), 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
2 instances · 2 analysts
· Δ adherence +5.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Select Start Case Entry
reference at this step: Open Study Section
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 76.8 (delta +5.7). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
2 instances · 1 analysts
· Δ adherence +5.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
View SUSAR Case
2
Create New Item
reference at this step: Open Study Section
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 76.8 (delta +5.7). 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 | 911 | 72.1% | 96.1% |
| collaboration.merck.com | 64 | 5.1% | 0.0% |
| Acrobat | 62 | 4.9% | 0.0% |
| usc-excel.officeapps.live.com | 49 | 3.9% | 0.0% |
| Microsoft Excel | 40 | 3.2% | 0.0% |
| Microsoft Word | 37 | 2.9% | 0.0% |
| Microsoft Teams | 36 | 2.8% | 0.0% |
| YoudaoDict | 13 | 1.0% | 0.0% |
| MEDDRABROWSERWIN | 8 | 0.6% | 0.0% |
| Phobos | 7 | 0.6% | 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 |
|---|---|---|
| collaboration.merck.com | usc-excel.officeapps.live.com | 39 |
| usc-excel.officeapps.live.com | collaboration.merck.com | 36 |
| Veeva Safety | Microsoft Excel | 28 |
| Veeva Safety | Acrobat | 22 |
| Acrobat | Microsoft Word | 21 |
| Acrobat | Veeva Safety | 20 |
| Veeva Safety | Microsoft Teams | 19 |
| Microsoft Word | Acrobat | 19 |
| Veeva Safety | collaboration.merck.com | 14 |
| Microsoft Excel | 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 2,972 events 91 cases 47.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
179×
pyzeClick
<span>
"Narrative"
123×
pyzeClick
<span>
"Adverse Events"
105×
pyzeClick
<a>
"Show more"
91×
pyzeClick
<span>
"Submissions & Distributions"
72×
pyzeClick
<button>
"Create"
66×
pyzeClick
<li>
"Narrative"
64×
pyzeClick
<span>
"Reference Numbers"
60×
Phobos DOM 526 events 49 cases 12.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
65×
pyzeClick
<button>
"Save Comments"
56×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
42×
pyzeClick
<button>
"Send to process"
33×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
32×
pyzeClick
<div>
"Please reply to “Request Information" using dropd"
30×
pyzeClick
<a>
"Cases"
23×
pyzeClick
<button>
"Advance"
23×
Microsoft Word Document 140 events 38 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 533164 (English) (26378749_0_4_342805)
659×
file
V-Narrative 534252 (English) (26399499_0_2_343143)
515×
file
V-Narrative 543823 (English) (26544362_0_2_343646)
514×
file
V-Narrative 540591 (English) (26495494_0_2_345212)
429×
file
V-Narrative 553016 (English) (26689302_0_2_348778)
392×
file
V-Narrative 528191 (English) (26300681_0_1_338349)
377×
file
V-Narrative 530932 (English) (26349278_0_2_339374)
366×
file
V-Narrative 555782 (English) (26733642_0_2_350941)
205×
Acrobat App 133 events 31 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
2,046 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.
collaboration.merck.com App 114 events 23 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
363 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 106 events 38 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,233 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 93 events 35 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
usc-excel.officeapps.live.com Document 70 events 10 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Outlook Message 31 events 18 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
apps.powerapps.com App 23 events 4 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.
0.17
Swivel Rate
events in non-primary app / total
0.92
Zero-Edit Visit Rate
inspection-only clicks
1.8
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 0.17
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Revisit Depth
value 1.80
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 |
|---|---|---|---|---|---|---|---|
| 2335072 | Analyst 13 | 31.1 | 7.2m | 27 | 0.96 | 0.00 | 12 |
| 2349107 | Analyst 13 | 31.9 | 5.3m | 32 | 0.94 | 0.00 | 14 |
| 2258699 | Analyst 13 | 32.7 | 0.9m | 22 | 0.91 | 0.00 | 10 |
| 2241891 | Analyst 01 | 34.0 | 20.6m | 45 | 0.89 | 0.00 | 20 |
| 2334048 | Analyst 12 | 36.6 | 1.9m | 25 | 0.80 | 0.00 | 10 |
Users on This Task
8 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
11
02
05
09
01
12
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 04 | 47 | 6.7 | 82.4 | Zero-Edit Visits | 0.00 | 0.00 | 796 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 04's Dominant Sequence
— 19 of 47 instances
1
View SUSAR Case
This analyst exhibits 26 distinct sequences across their 47 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 11 | 2 | 2.0 | 82.0 | Zero-Edit Visits | 0.00 | 0.00 | 228 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 11's Dominant Sequence
— 1 of 2 instances
1
View SUSAR Case
This analyst exhibits 2 distinct sequences across their 2 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 53 | 6.6 | 78.4 | Zero-Edit Visits | 0.13 | 0.01 | 96 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 02's Dominant Sequence
— 20 of 53 instances
1
View SUSAR Case
This analyst exhibits 32 distinct sequences across their 53 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 8 | 1.6 | 77.1 | Zero-Edit Visits | 0.21 | 0.00 | 10 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 05's Dominant Sequence
— 3 of 8 instances
1
View SUSAR Case
This analyst exhibits 6 distinct sequences across their 8 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 84 | 7.0 | 76.6 | Zero-Edit Visits | 0.18 | 0.00 | 2,065 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 09's Dominant Sequence
— 23 of 84 instances
1
View SUSAR Case
This analyst exhibits 56 distinct sequences across their 84 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 15 | 2.1 | 75.5 | Zero-Edit Visits | 0.17 | 0.00 | 84 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 01's Dominant Sequence
— 3 of 15 instances
1
View SUSAR Case
This analyst exhibits 12 distinct sequences across their 15 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 21 | 5.2 | 74.1 | Zero-Edit Visits | 0.24 | 0.00 | 128 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 12's Dominant Sequence
— 7 of 21 instances
1
View SUSAR Case
This analyst exhibits 14 distinct sequences across their 21 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 39 | 4.3 | 69.8 | Zero-Edit Visits | 0.34 | 0.00 | 1,362 |
|
Reference SoP
1
View SUSAR Case
2
Open Study Section
Analyst 13's Dominant Sequence
— 10 of 39 instances
1
View SUSAR Case
This analyst exhibits 29 distinct sequences across their 39 instances; the dominant one is shown.
|
||||||||