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 63 instances
Generate Narrative (AI-assisted)
How 9 analysts execute this task across 55 cases — auto-generated Standard Operating Procedure with adherence scoring.
79.9
Avg Adherence Score (0-100)
Median 82.5 · Range 56.0–90.0
1.7 min
Avg Task Duration
0.4 min median
2.6
Avg Sub-Steps per Instance
50.8% have multiple sub-steps
1.6%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
1.6% of instances follow this exactly
Generate Narrative (AI-assisted) — Mined SoP
The most common sub-step sequence observed across 63 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Generate Narrative from Outline
0.0s
Veeva Safety
100%
2
Open Details Section
0.0s
Veeva Safety
3%
3
Navigate to Cases
0.0s
Veeva Safety
2%
4
gpteal.merck.com
0.0s
gpteal.merck.com
2%
5
Microsoft Word
0.0s
Microsoft Word
2%
6
Select Patient
0.0s
Veeva Safety
2%
7
Microsoft Word
0.0s
Microsoft Word
2%
Variant Explorer
Reference
Generate Narrative (AI-assisted) 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 Generate Narrative (AI-assisted).
Genuine Complexity
Complex-Case Execution Path
31 instances · 8 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
×
Open Details Section
skipped — this reference step was not executed
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 7 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
2 instances · 2 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Preview Japanese Narrative
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 7 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
2 instances · 2 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Show More Details
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 7 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.
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Select Documents
reference at this step: Open Details Section
3
View Validation Results Count
reference at this step: Navigate to Cases
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
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 1 cases.
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence -7.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
View Site Narrative
reference at this step: Open Details Section
3
View Case Description
reference at this step: Navigate to Cases
4
Show More Details
reference at this step: gpteal.merck.com
5
View Investigator Information
reference at this step: Microsoft Word
6
SnippingTool
reference at this step: Select Patient
7
Show More Details
reference at this step: Microsoft Word
8
Microsoft PowerPoint
extra step (reference ends at step 7)
Reviewer Sweep — a viewing-heavy pattern with 8.0 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 1 cases.
Genuine Complexity
Complex-Case Execution Path
1 instances · 1 analysts
· Δ adherence -5.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
View Validation Results
reference at this step: Open Details Section
3
Select Patient
reference at this step: Navigate to Cases
4
Select Adverse Events
reference at this step: gpteal.merck.com
5
Microsoft Outlook
reference at this step: Microsoft Word
6
Microsoft Teams
reference at this step: Select Patient
7
View Validation Results
reference at this step: Microsoft Word
8
Select Patient
extra step (reference ends at step 7)
9
Cancel
extra step (reference ends at step 7)
10
Acrobat
extra step (reference ends at step 7)
11
Select Drug History
extra step (reference ends at step 7)
12
Select Medical History
extra step (reference ends at step 7)
13
Create New Item
extra step (reference ends at step 7)
14
Navigate to Cases
extra step (reference ends at step 7)
Genuine Complexity — this sequence differs from the reference path (14 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 74.6 vs reference 79.9 (delta -5.3). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Teams
1 instances · 1 analysts
· Δ adherence -12.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Microsoft Teams
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Cross-App Swivel — 1 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.
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence -4.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
View Site Narrative
reference at this step: Open Details Section
3
Preview German Narrative
reference at this step: Navigate to Cases
4
View Site Narrative
reference at this step: gpteal.merck.com
5
View Case Description
reference at this step: Microsoft Word
6
Microsoft Word
reference at this step: Select Patient
×
Microsoft Word
skipped — this reference step was not executed
Reviewer Sweep — a viewing-heavy pattern with 6.0 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 1 cases.
Cross-App Swivel
Cross-App Lookup to Microsoft Outlook
1 instances · 1 analysts
· Δ adherence -12.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Microsoft Outlook
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Cross-App Swivel — 1 analysts pivot into Microsoft Outlook 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
1 instances · 1 analysts
· Δ adherence +7.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Update Case Narrative
reference at this step: Open Details Section
3
Select List Item
reference at this step: Navigate to Cases
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (3 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 87.5 vs reference 79.9 (delta +7.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
1 instances · 1 analysts
· Δ adherence +10.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Complete Link
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 90.0 vs reference 79.9 (delta +10.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Select Medical History
reference at this step: Open Details Section
3
Create New Item
reference at this step: Navigate to Cases
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
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 1 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 | 135 | 81.8% | 100.0% |
| Microsoft Teams | 7 | 4.2% | 0.0% |
| Acrobat | 5 | 3.0% | 0.0% |
| Microsoft Word | 4 | 2.4% | 0.0% |
| TextInputHost | 2 | 1.2% | 0.0% |
| Microsoft Outlook | 2 | 1.2% | 0.0% |
| gpteal.merck.com | 2 | 1.2% | 0.0% |
| Microsoft Excel | 2 | 1.2% | 0.0% |
| newtab | 1 | 0.6% | 0.0% |
| dtgpteal.merck.com | 1 | 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 |
|---|---|---|
| Veeva Safety | Microsoft Teams | 4 |
| Veeva Safety | Acrobat | 4 |
| Acrobat | Veeva Safety | 3 |
| Microsoft Teams | Veeva Safety | 3 |
| Microsoft Teams | TextInputHost | 2 |
| Veeva Safety | Microsoft Excel | 2 |
| Veeva Safety | Microsoft Outlook | 2 |
| Microsoft Excel | Veeva Safety | 2 |
| Veeva Safety | Microsoft Word | 2 |
| Acrobat | gpteal.merck.com | 1 |
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 1,553 events 55 cases 23.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
145×
pyzeClick
<a>
"Show more"
75×
pyzeClick
<span>
"Generate Narrative from Outline"
65×
pyzeClick
<li>
"Narrative"
60×
pyzeClick
<span>
"Narrative"
48×
pyzeClick
<button>
"Create"
40×
pyzeClick
<span>
"Submissions & Distributions"
34×
pyzeClick
<li>
"Unknown"
33×
Phobos DOM 156 events 22 cases 4.6 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
21×
pyzeClick
<button>
"Advance"
15×
pyzeClick
<button>
"Save Comments"
13×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
9×
pyzeClick
<button>
"Send to process"
6×
inputld
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
5×
pyzeClick
<a>
"Cases"
5×
pyzeClick
<div>
"Case version number · First Inbox Item · Compound · Sel"
4×
Microsoft Word Document 54 events 20 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 525025 (English) (26253159_0_3_342205)
211×
file
V-Narrative 542512 (English) (26522671_0_2_344761)
189×
file
V-Narrative 531284 (English) (26355806_0_2_340584)
156×
file
V-Narrative 523170 (English) (26221379_0_2_340465)
144×
file
V-Narrative 433845 (English) (24804813_0_2_345392)
137×
file
V-Narrative 530932 (English) (26349278_0_2_339374)
118×
file
V-Narrative 537919 (English) (26452534_0_2_341686)
99×
file
V-Narrative 498632 (English) (25800349_0_2_340329)
87×
Microsoft Teams App 44 events 21 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
534 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Acrobat App 43 events 16 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
390 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 19 events 10 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
102×
file
All PV Smart Submissions 28-NOV-2025 PM
12×
file
records_as_of_2025_12_01_EST
12×
file
Case tracking list
9×
file
Priorización
5×
ONENOTE Document 9 events 5 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Outlook Message 9 events 7 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
gpteal.merck.com DOM 6 events 3 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
olk Message 6 events 4 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.09
Swivel Rate
events in non-primary app / total
0.98
Zero-Edit Visit Rate
inspection-only clicks
1.11
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.98
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 0.98
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.11
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.09
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2357099 | Analyst 12 | 56.0 | 1.8m | 5 | 0.80 | 0.00 | 2 |
| 2347303 | Analyst 02 | 62.5 | 24.8m | 8 | 0.50 | 0.00 | 3 |
| 2351484 | Analyst 02 | 67.5 | 0.9m | 2 | 0.50 | 0.00 | 1 |
| 2354665 | Analyst 02 | 67.5 | 0.5m | 2 | 0.50 | 0.00 | 1 |
| 2351809 | Analyst 01 | 67.5 | 0.3m | 2 | 0.50 | 0.00 | 1 |
Users on This Task
9 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) →
05
14
13
04
15
12
01
09
02
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 05 | 4 | 2.0 | 85.6 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 05's Dominant Sequence
— 2 of 4 instances
1
Generate Narrative from Outline
This analyst exhibits 3 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 5 | 2.5 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 5 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 14's Dominant Sequence
— 3 of 5 instances
1
Generate Narrative from Outline
This analyst exhibits 3 distinct sequences across their 5 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 13's Dominant Sequence
— 1 of 1 instances
1
Generate Narrative from Outline
|
||||||||
| ▸ | Analyst 04 | 15 | 2.1 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 10 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 04's Dominant Sequence
— 10 of 15 instances
1
Generate Narrative from Outline
This analyst exhibits 6 distinct sequences across their 15 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 3 | 3.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 15's Dominant Sequence
— 1 of 3 instances
1
Generate Narrative from Outline
2
Select Case Assessment Results
· reference: Open Details Section
This analyst exhibits 3 distinct sequences across their 3 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 8 | 1.6 | 79.8 | Zero-Edit Visits | 0.10 | 0.00 | 10 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 12's Dominant Sequence
— 4 of 8 instances
1
Generate Narrative from Outline
This analyst exhibits 5 distinct sequences across their 8 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 13 | 1.9 | 78.4 | Zero-Edit Visits | 0.14 | 0.00 | 6 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 01's Dominant Sequence
— 8 of 13 instances
1
Generate Narrative from Outline
This analyst exhibits 6 distinct sequences across their 13 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 7 | 1.8 | 78.3 | Zero-Edit Visits | 0.13 | 0.00 | 24 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 09's Dominant Sequence
— 2 of 7 instances
1
Generate Narrative from Outline
This analyst exhibits 6 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 7 | 1.2 | 71.7 | Zero-Edit Visits | 0.30 | 0.00 | 46 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 02's Dominant Sequence
— 1 of 7 instances
1
Generate Narrative from Outline
2
View Site Narrative
· reference: Open Details Section
3
View Case Description
· reference: Navigate to Cases
4
Show More Details
· reference: gpteal.merck.com
5
View Investigator Information
· reference: Microsoft Word
6
SnippingTool
· reference: Select Patient
7
Show More Details
· reference: Microsoft Word
8
Microsoft PowerPoint
This analyst exhibits 7 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||
79.4
Avg Adherence Score (0-100)
Median 82.5 · Range 48.9–90.0
2.1 min
Avg Task Duration
0.5 min median
2.9
Avg Sub-Steps per Instance
50.8% have multiple sub-steps
1.6%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
1.6% of instances follow this exactly
Generate Narrative (AI-assisted) — Mined SoP
The most common sub-step sequence observed across 63 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Generate Narrative from Outline
0.0s
Veeva Safety
100%
2
Open Details Section
0.0s
Veeva Safety
3%
3
Navigate to Cases
0.0s
Veeva Safety
2%
4
gpteal.merck.com
0.0s
gpteal.merck.com
3%
5
Microsoft Word
0.0s
Microsoft Word
3%
6
Select Patient
0.0s
Veeva Safety
2%
7
Microsoft Word
0.0s
Microsoft Word
3%
Variant Explorer
Reference
Generate Narrative (AI-assisted) 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 Generate Narrative (AI-assisted).
Genuine Complexity
Complex-Case Execution Path
31 instances · 8 analysts
· Δ adherence +3.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
×
Open Details Section
skipped — this reference step was not executed
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 82.5 vs reference 79.4 (delta +3.1). 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 +3.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Preview Japanese Narrative
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 82.5 vs reference 79.4 (delta +3.1). 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 +3.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Show More Details
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 82.5 vs reference 79.4 (delta +3.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence +3.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Select Documents
reference at this step: Open Details Section
3
View Validation Results Count
reference at this step: Navigate to Cases
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
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 1 cases.
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence -6.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
View Site Narrative
reference at this step: Open Details Section
3
View Case Description
reference at this step: Navigate to Cases
4
Show More Details
reference at this step: gpteal.merck.com
5
View Investigator Information
reference at this step: Microsoft Word
6
SnippingTool
reference at this step: Select Patient
7
Show More Details
reference at this step: Microsoft Word
8
Microsoft PowerPoint
extra step (reference ends at step 7)
Reviewer Sweep — a viewing-heavy pattern with 8.0 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 1 cases.
Genuine Complexity
Complex-Case Execution Path
1 instances · 1 analysts
· Δ adherence -4.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
View Validation Results
reference at this step: Open Details Section
3
Select Patient
reference at this step: Navigate to Cases
4
Select Adverse Events
reference at this step: gpteal.merck.com
5
Microsoft Outlook
reference at this step: Microsoft Word
6
Microsoft Teams
reference at this step: Select Patient
7
View Validation Results
reference at this step: Microsoft Word
8
Select Patient
extra step (reference ends at step 7)
9
Cancel
extra step (reference ends at step 7)
10
Acrobat
extra step (reference ends at step 7)
11
Select Drug History
extra step (reference ends at step 7)
12
Select Medical History
extra step (reference ends at step 7)
13
Create New Item
extra step (reference ends at step 7)
14
Navigate to Cases
extra step (reference ends at step 7)
Genuine Complexity — this sequence differs from the reference path (14 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 74.6 vs reference 79.4 (delta -4.8). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Teams
1 instances · 1 analysts
· Δ adherence -11.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Microsoft Teams
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Cross-App Swivel — 1 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.
Cross-App Swivel
Cross-App Lookup to Microsoft Word
1 instances · 1 analysts
· Δ adherence -13.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
View Site Narrative
reference at this step: Open Details Section
3
Preview German Narrative
reference at this step: Navigate to Cases
4
View Site Narrative
reference at this step: gpteal.merck.com
5
View Case Description
reference at this step: Microsoft Word
6
Microsoft Word
reference at this step: Select Patient
7
Select Products
reference at this step: Microsoft Word
8
Microsoft Word
extra step (reference ends at step 7)
9
Acrobat
extra step (reference ends at step 7)
10
Microsoft Word
extra step (reference ends at step 7)
Cross-App Swivel — 1 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 Outlook
1 instances · 1 analysts
· Δ adherence -11.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Microsoft Outlook
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Cross-App Swivel — 1 analysts pivot into Microsoft Outlook 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
1 instances · 1 analysts
· Δ adherence +8.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Update Case Narrative
reference at this step: Open Details Section
3
Select List Item
reference at this step: Navigate to Cases
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (3 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 87.5 vs reference 79.4 (delta +8.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
1 instances · 1 analysts
· Δ adherence +10.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Complete Link
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 90.0 vs reference 79.4 (delta +10.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence +3.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Select Medical History
reference at this step: Open Details Section
3
Create New Item
reference at this step: Navigate to Cases
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
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 1 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 | 138 | 76.2% | 100.0% |
| Microsoft Word | 11 | 6.1% | 0.0% |
| Microsoft Teams | 7 | 3.9% | 0.0% |
| Acrobat | 6 | 3.3% | 0.0% |
| ONENOTE | 4 | 2.2% | 0.0% |
| gpteal.merck.com | 3 | 1.7% | 0.0% |
| Microsoft Excel | 3 | 1.7% | 0.0% |
| Microsoft Outlook | 2 | 1.1% | 0.0% |
| TextInputHost | 2 | 1.1% | 0.0% |
| SnippingTool | 1 | 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 |
|---|---|---|
| Veeva Safety | Acrobat | 4 |
| Veeva Safety | Microsoft Teams | 4 |
| Veeva Safety | Microsoft Word | 3 |
| Veeva Safety | Microsoft Excel | 3 |
| Microsoft Word | Veeva Safety | 3 |
| Microsoft Teams | Veeva Safety | 3 |
| ONENOTE | Microsoft Word | 3 |
| Acrobat | Veeva Safety | 3 |
| Microsoft Word | ONENOTE | 3 |
| Veeva Safety | gpteal.merck.com | 2 |
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 1,553 events 55 cases 23.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
145×
pyzeClick
<a>
"Show more"
75×
pyzeClick
<span>
"Generate Narrative from Outline"
65×
pyzeClick
<li>
"Narrative"
60×
pyzeClick
<span>
"Narrative"
48×
pyzeClick
<button>
"Create"
40×
pyzeClick
<span>
"Submissions & Distributions"
34×
pyzeClick
<li>
"Unknown"
33×
Phobos DOM 156 events 22 cases 4.6 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
21×
pyzeClick
<button>
"Advance"
15×
pyzeClick
<button>
"Save Comments"
13×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
9×
pyzeClick
<button>
"Send to process"
6×
inputld
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
5×
pyzeClick
<a>
"Cases"
5×
pyzeClick
<div>
"Case version number · First Inbox Item · Compound · Sel"
4×
Microsoft Word Document 54 events 20 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 525025 (English) (26253159_0_3_342205)
211×
file
V-Narrative 542512 (English) (26522671_0_2_344761)
189×
file
V-Narrative 531284 (English) (26355806_0_2_340584)
156×
file
V-Narrative 523170 (English) (26221379_0_2_340465)
144×
file
V-Narrative 433845 (English) (24804813_0_2_345392)
137×
file
V-Narrative 530932 (English) (26349278_0_2_339374)
118×
file
V-Narrative 537919 (English) (26452534_0_2_341686)
99×
file
V-Narrative 498632 (English) (25800349_0_2_340329)
87×
Microsoft Teams App 44 events 21 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
534 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Acrobat App 43 events 16 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
390 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 19 events 10 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
102×
file
All PV Smart Submissions 28-NOV-2025 PM
12×
file
records_as_of_2025_12_01_EST
12×
file
Case tracking list
9×
file
Priorización
5×
Microsoft Outlook Message 9 events 7 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
ONENOTE Document 9 events 5 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
gpteal.merck.com DOM 6 events 3 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
olk Message 6 events 4 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.10
Swivel Rate
events in non-primary app / total
0.98
Zero-Edit Visit Rate
inspection-only clicks
1.19
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.98
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 0.98
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.19
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.10
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 |
|---|---|---|---|---|---|---|---|
| 2351809 | Analyst 01 | 48.9 | 16.8m | 14 | 0.79 | 0.00 | 5 |
| 2357099 | Analyst 12 | 56.0 | 1.8m | 5 | 0.80 | 0.00 | 2 |
| 2347303 | Analyst 02 | 62.5 | 24.8m | 8 | 0.50 | 0.00 | 3 |
| 2352746 | Analyst 02 | 65.5 | 12.7m | 10 | 0.40 | 0.00 | 3 |
| 2351484 | Analyst 02 | 67.5 | 0.9m | 2 | 0.50 | 0.00 | 1 |
Users on This Task
9 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) →
05
04
15
14
13
12
09
01
02
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 05 | 4 | 2.0 | 85.6 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 05's Dominant Sequence
— 2 of 4 instances
1
Generate Narrative from Outline
This analyst exhibits 3 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 15 | 2.1 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 10 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 04's Dominant Sequence
— 10 of 15 instances
1
Generate Narrative from Outline
This analyst exhibits 6 distinct sequences across their 15 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 3 | 3.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 15's Dominant Sequence
— 1 of 3 instances
1
Generate Narrative from Outline
2
Select Case Assessment Results
· reference: Open Details Section
This analyst exhibits 3 distinct sequences across their 3 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 5 | 2.5 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 5 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 14's Dominant Sequence
— 3 of 5 instances
1
Generate Narrative from Outline
This analyst exhibits 3 distinct sequences across their 5 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 13's Dominant Sequence
— 1 of 1 instances
1
Generate Narrative from Outline
|
||||||||
| ▸ | Analyst 12 | 8 | 1.6 | 79.8 | Zero-Edit Visits | 0.10 | 0.00 | 10 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 12's Dominant Sequence
— 4 of 8 instances
1
Generate Narrative from Outline
This analyst exhibits 5 distinct sequences across their 8 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 7 | 1.8 | 78.3 | Zero-Edit Visits | 0.13 | 0.00 | 24 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 09's Dominant Sequence
— 2 of 7 instances
1
Generate Narrative from Outline
This analyst exhibits 6 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 13 | 1.9 | 77.0 | Zero-Edit Visits | 0.16 | 0.00 | 23 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 01's Dominant Sequence
— 8 of 13 instances
1
Generate Narrative from Outline
This analyst exhibits 6 distinct sequences across their 13 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 7 | 1.2 | 70.4 | Zero-Edit Visits | 0.34 | 0.00 | 56 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 02's Dominant Sequence
— 1 of 7 instances
1
Generate Narrative from Outline
2
View Site Narrative
· reference: Open Details Section
3
View Case Description
· reference: Navigate to Cases
4
Show More Details
· reference: gpteal.merck.com
5
View Investigator Information
· reference: Microsoft Word
6
SnippingTool
· reference: Select Patient
7
Show More Details
· reference: Microsoft Word
8
Microsoft PowerPoint
This analyst exhibits 7 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||
79.9
Avg Adherence Score (0-100)
Median 82.5 · Range 56.0–90.0
1.7 min
Avg Task Duration
0.4 min median
2.6
Avg Sub-Steps per Instance
50.8% have multiple sub-steps
1.6%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
1.6% of instances follow this exactly
Generate Narrative (AI-assisted) — Mined SoP
The most common sub-step sequence observed across 63 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Generate Narrative from Outline
0.0s
Veeva Safety
100%
2
Open Details Section
0.0s
Veeva Safety
3%
3
Navigate to Cases
0.0s
Veeva Safety
2%
4
gpteal.merck.com
0.0s
gpteal.merck.com
2%
5
Microsoft Word
0.0s
Microsoft Word
2%
6
Select Patient
0.0s
Veeva Safety
2%
7
Microsoft Word
0.0s
Microsoft Word
2%
Variant Explorer
Reference
Generate Narrative (AI-assisted) 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 Generate Narrative (AI-assisted).
Genuine Complexity
Complex-Case Execution Path
31 instances · 8 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
×
Open Details Section
skipped — this reference step was not executed
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 7 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
2 instances · 2 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Preview Japanese Narrative
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 7 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
2 instances · 2 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Show More Details
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 7 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.
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Select Documents
reference at this step: Open Details Section
3
View Validation Results Count
reference at this step: Navigate to Cases
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
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 1 cases.
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence -7.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
View Site Narrative
reference at this step: Open Details Section
3
View Case Description
reference at this step: Navigate to Cases
4
Show More Details
reference at this step: gpteal.merck.com
5
View Investigator Information
reference at this step: Microsoft Word
6
SnippingTool
reference at this step: Select Patient
7
Show More Details
reference at this step: Microsoft Word
8
Microsoft PowerPoint
extra step (reference ends at step 7)
Reviewer Sweep — a viewing-heavy pattern with 8.0 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 1 cases.
Genuine Complexity
Complex-Case Execution Path
1 instances · 1 analysts
· Δ adherence -5.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
View Validation Results
reference at this step: Open Details Section
3
Select Patient
reference at this step: Navigate to Cases
4
Select Adverse Events
reference at this step: gpteal.merck.com
5
Microsoft Outlook
reference at this step: Microsoft Word
6
Microsoft Teams
reference at this step: Select Patient
7
View Validation Results
reference at this step: Microsoft Word
8
Select Patient
extra step (reference ends at step 7)
9
Cancel
extra step (reference ends at step 7)
10
Acrobat
extra step (reference ends at step 7)
11
Select Drug History
extra step (reference ends at step 7)
12
Select Medical History
extra step (reference ends at step 7)
13
Create New Item
extra step (reference ends at step 7)
14
Navigate to Cases
extra step (reference ends at step 7)
Genuine Complexity — this sequence differs from the reference path (14 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 74.6 vs reference 79.9 (delta -5.3). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Teams
1 instances · 1 analysts
· Δ adherence -12.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Microsoft Teams
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Cross-App Swivel — 1 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.
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence -4.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
View Site Narrative
reference at this step: Open Details Section
3
Preview German Narrative
reference at this step: Navigate to Cases
4
View Site Narrative
reference at this step: gpteal.merck.com
5
View Case Description
reference at this step: Microsoft Word
6
Microsoft Word
reference at this step: Select Patient
×
Microsoft Word
skipped — this reference step was not executed
Reviewer Sweep — a viewing-heavy pattern with 6.0 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 1 cases.
Cross-App Swivel
Cross-App Lookup to Microsoft Outlook
1 instances · 1 analysts
· Δ adherence -12.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Microsoft Outlook
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Cross-App Swivel — 1 analysts pivot into Microsoft Outlook 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
1 instances · 1 analysts
· Δ adherence +7.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Update Case Narrative
reference at this step: Open Details Section
3
Select List Item
reference at this step: Navigate to Cases
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (3 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 87.5 vs reference 79.9 (delta +7.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
1 instances · 1 analysts
· Δ adherence +10.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Complete Link
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 90.0 vs reference 79.9 (delta +10.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence +2.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Select Medical History
reference at this step: Open Details Section
3
Create New Item
reference at this step: Navigate to Cases
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
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 1 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 | 135 | 81.8% | 100.0% |
| Microsoft Teams | 7 | 4.2% | 0.0% |
| Acrobat | 5 | 3.0% | 0.0% |
| Microsoft Word | 4 | 2.4% | 0.0% |
| TextInputHost | 2 | 1.2% | 0.0% |
| Microsoft Outlook | 2 | 1.2% | 0.0% |
| gpteal.merck.com | 2 | 1.2% | 0.0% |
| Microsoft Excel | 2 | 1.2% | 0.0% |
| newtab | 1 | 0.6% | 0.0% |
| dtgpteal.merck.com | 1 | 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 |
|---|---|---|
| Veeva Safety | Acrobat | 4 |
| Veeva Safety | Microsoft Teams | 4 |
| Microsoft Teams | Veeva Safety | 3 |
| Acrobat | Veeva Safety | 3 |
| Microsoft Excel | Veeva Safety | 2 |
| Veeva Safety | Microsoft Outlook | 2 |
| Veeva Safety | Microsoft Word | 2 |
| Microsoft Teams | TextInputHost | 2 |
| Veeva Safety | Microsoft Excel | 2 |
| Microsoft Word | Veeva Safety | 1 |
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 1,553 events 55 cases 23.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
145×
pyzeClick
<a>
"Show more"
75×
pyzeClick
<span>
"Generate Narrative from Outline"
65×
pyzeClick
<li>
"Narrative"
60×
pyzeClick
<span>
"Narrative"
48×
pyzeClick
<button>
"Create"
40×
pyzeClick
<span>
"Submissions & Distributions"
34×
pyzeClick
<li>
"Unknown"
33×
Phobos DOM 156 events 22 cases 4.6 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
21×
pyzeClick
<button>
"Advance"
15×
pyzeClick
<button>
"Save Comments"
13×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
9×
pyzeClick
<button>
"Send to process"
6×
pyzeClick
<a>
"Cases"
5×
inputld
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
5×
pyzeClick
<button>
"Send to Medical Review"
4×
Microsoft Word Document 54 events 20 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 525025 (English) (26253159_0_3_342205)
211×
file
V-Narrative 542512 (English) (26522671_0_2_344761)
189×
file
V-Narrative 531284 (English) (26355806_0_2_340584)
156×
file
V-Narrative 523170 (English) (26221379_0_2_340465)
144×
file
V-Narrative 433845 (English) (24804813_0_2_345392)
137×
file
V-Narrative 530932 (English) (26349278_0_2_339374)
118×
file
V-Narrative 537919 (English) (26452534_0_2_341686)
99×
file
V-Narrative 498632 (English) (25800349_0_2_340329)
87×
Microsoft Teams App 44 events 21 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
534 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Acrobat App 43 events 16 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
390 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 19 events 10 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
102×
file
All PV Smart Submissions 28-NOV-2025 PM
12×
file
records_as_of_2025_12_01_EST
12×
file
Case tracking list
9×
file
Priorización
5×
ONENOTE Document 9 events 5 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Outlook Message 9 events 7 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
gpteal.merck.com DOM 6 events 3 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
olk Message 6 events 4 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.09
Swivel Rate
events in non-primary app / total
0.98
Zero-Edit Visit Rate
inspection-only clicks
1.11
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.98
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 0.98
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.11
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.09
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.00
Time outside primary app dominates this score. Identify which secondary app(s) and whether their function can be in-lined.
Lowest-Adherence Instances
The 5 task instances that scored lowest. These are the concrete cases worth investigating to understand the worst patterns.
| Case | Analyst | Score | Duration | Sub-Steps | Swivel | X-App | Revisit |
|---|---|---|---|---|---|---|---|
| 2357099 | Analyst 12 | 56.0 | 1.8m | 5 | 0.80 | 0.00 | 2 |
| 2347303 | Analyst 02 | 62.5 | 24.8m | 8 | 0.50 | 0.00 | 3 |
| 2351809 | Analyst 01 | 67.5 | 0.3m | 2 | 0.50 | 0.00 | 1 |
| 2352839 | Analyst 01 | 67.5 | 1.6m | 6 | 0.50 | 0.00 | 1 |
| 2354665 | Analyst 02 | 67.5 | 0.5m | 2 | 0.50 | 0.00 | 1 |
Users on This Task
9 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) →
05
04
15
14
13
12
01
09
02
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 05 | 4 | 2.0 | 85.6 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 05's Dominant Sequence
— 2 of 4 instances
1
Generate Narrative from Outline
This analyst exhibits 3 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 15 | 2.1 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 10 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 04's Dominant Sequence
— 10 of 15 instances
1
Generate Narrative from Outline
This analyst exhibits 6 distinct sequences across their 15 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 3 | 3.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 15's Dominant Sequence
— 1 of 3 instances
1
Generate Narrative from Outline
2
Select Case Assessment Results
· reference: Open Details Section
This analyst exhibits 3 distinct sequences across their 3 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 5 | 2.5 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 5 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 14's Dominant Sequence
— 3 of 5 instances
1
Generate Narrative from Outline
This analyst exhibits 3 distinct sequences across their 5 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 13's Dominant Sequence
— 1 of 1 instances
1
Generate Narrative from Outline
|
||||||||
| ▸ | Analyst 12 | 8 | 1.6 | 79.8 | Zero-Edit Visits | 0.10 | 0.00 | 10 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 12's Dominant Sequence
— 4 of 8 instances
1
Generate Narrative from Outline
This analyst exhibits 5 distinct sequences across their 8 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 13 | 1.9 | 78.4 | Zero-Edit Visits | 0.14 | 0.00 | 6 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 01's Dominant Sequence
— 8 of 13 instances
1
Generate Narrative from Outline
This analyst exhibits 6 distinct sequences across their 13 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 7 | 1.8 | 78.3 | Zero-Edit Visits | 0.13 | 0.00 | 24 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 09's Dominant Sequence
— 2 of 7 instances
1
Generate Narrative from Outline
This analyst exhibits 6 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 7 | 1.2 | 71.7 | Zero-Edit Visits | 0.30 | 0.00 | 46 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 02's Dominant Sequence
— 1 of 7 instances
1
Generate Narrative from Outline
2
View Site Narrative
· reference: Open Details Section
3
View Case Description
· reference: Navigate to Cases
4
Show More Details
· reference: gpteal.merck.com
5
View Investigator Information
· reference: Microsoft Word
6
SnippingTool
· reference: Select Patient
7
Show More Details
· reference: Microsoft Word
8
Microsoft PowerPoint
This analyst exhibits 7 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||
79.4
Avg Adherence Score (0-100)
Median 82.5 · Range 48.9–90.0
2.1 min
Avg Task Duration
0.5 min median
2.9
Avg Sub-Steps per Instance
50.8% have multiple sub-steps
1.6%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
1.6% of instances follow this exactly
Generate Narrative (AI-assisted) — Mined SoP
The most common sub-step sequence observed across 63 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Generate Narrative from Outline
0.0s
Veeva Safety
100%
2
Open Details Section
0.0s
Veeva Safety
3%
3
Navigate to Cases
0.0s
Veeva Safety
2%
4
gpteal.merck.com
0.0s
gpteal.merck.com
3%
5
Microsoft Word
0.0s
Microsoft Word
3%
6
Select Patient
0.0s
Veeva Safety
2%
7
Microsoft Word
0.0s
Microsoft Word
3%
Variant Explorer
Reference
Generate Narrative (AI-assisted) 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 Generate Narrative (AI-assisted).
Genuine Complexity
Complex-Case Execution Path
31 instances · 8 analysts
· Δ adherence +3.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
×
Open Details Section
skipped — this reference step was not executed
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 82.5 vs reference 79.4 (delta +3.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Linked Remediation Opportunities
Genuine Complexity
Complex-Case Execution Path
2 instances · 2 analysts
· Δ adherence +3.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Preview Japanese Narrative
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 82.5 vs reference 79.4 (delta +3.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Linked Remediation Opportunities
Genuine Complexity
Complex-Case Execution Path
2 instances · 2 analysts
· Δ adherence +3.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Show More Details
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 82.5 vs reference 79.4 (delta +3.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Linked Remediation Opportunities
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence +3.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Select Documents
reference at this step: Open Details Section
3
View Validation Results Count
reference at this step: Navigate to Cases
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
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 1 cases.
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence -6.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
View Site Narrative
reference at this step: Open Details Section
3
View Case Description
reference at this step: Navigate to Cases
4
Show More Details
reference at this step: gpteal.merck.com
5
View Investigator Information
reference at this step: Microsoft Word
6
SnippingTool
reference at this step: Select Patient
7
Show More Details
reference at this step: Microsoft Word
8
Microsoft PowerPoint
extra step (reference ends at step 7)
Reviewer Sweep — a viewing-heavy pattern with 8.0 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 1 cases.
Genuine Complexity
Complex-Case Execution Path
1 instances · 1 analysts
· Δ adherence -4.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
View Validation Results
reference at this step: Open Details Section
3
Select Patient
reference at this step: Navigate to Cases
4
Select Adverse Events
reference at this step: gpteal.merck.com
5
Microsoft Outlook
reference at this step: Microsoft Word
6
Microsoft Teams
reference at this step: Select Patient
7
View Validation Results
reference at this step: Microsoft Word
8
Select Patient
extra step (reference ends at step 7)
9
Cancel
extra step (reference ends at step 7)
10
Acrobat
extra step (reference ends at step 7)
11
Select Drug History
extra step (reference ends at step 7)
12
Select Medical History
extra step (reference ends at step 7)
13
Create New Item
extra step (reference ends at step 7)
14
Navigate to Cases
extra step (reference ends at step 7)
Genuine Complexity — this sequence differs from the reference path (14 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 74.6 vs reference 79.4 (delta -4.8). 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
1 instances · 1 analysts
· Δ adherence -11.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Microsoft Teams
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Cross-App Swivel — 1 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.
Cross-App Swivel
Cross-App Lookup to Microsoft Word
1 instances · 1 analysts
· Δ adherence -13.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
View Site Narrative
reference at this step: Open Details Section
3
Preview German Narrative
reference at this step: Navigate to Cases
4
View Site Narrative
reference at this step: gpteal.merck.com
5
View Case Description
reference at this step: Microsoft Word
6
Microsoft Word
reference at this step: Select Patient
7
Select Products
reference at this step: Microsoft Word
8
Microsoft Word
extra step (reference ends at step 7)
9
Acrobat
extra step (reference ends at step 7)
10
Microsoft Word
extra step (reference ends at step 7)
Cross-App Swivel — 1 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 Outlook
1 instances · 1 analysts
· Δ adherence -11.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Microsoft Outlook
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Cross-App Swivel — 1 analysts pivot into Microsoft Outlook 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
1 instances · 1 analysts
· Δ adherence +8.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Update Case Narrative
reference at this step: Open Details Section
3
Select List Item
reference at this step: Navigate to Cases
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (3 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 87.5 vs reference 79.4 (delta +8.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Linked Remediation Opportunities
Genuine Complexity
Complex-Case Execution Path
1 instances · 1 analysts
· Δ adherence +10.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Complete Link
reference at this step: Open Details Section
×
Navigate to Cases
skipped — this reference step was not executed
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 7 in reference) but shows no strong pain-point signals: adherence score 90.0 vs reference 79.4 (delta +10.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Linked Remediation Opportunities
Reviewer Sweep
Reviewer Verification Pass
1 instances · 1 analysts
· Δ adherence +3.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Generate Narrative from Outline
2
Select Medical History
reference at this step: Open Details Section
3
Create New Item
reference at this step: Navigate to Cases
×
gpteal.merck.com
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
×
Select Patient
skipped — this reference step was not executed
×
Microsoft Word
skipped — this reference step was not executed
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 1 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 | 138 | 76.2% | 100.0% |
| Microsoft Word | 11 | 6.1% | 0.0% |
| Microsoft Teams | 7 | 3.9% | 0.0% |
| Acrobat | 6 | 3.3% | 0.0% |
| ONENOTE | 4 | 2.2% | 0.0% |
| gpteal.merck.com | 3 | 1.7% | 0.0% |
| Microsoft Excel | 3 | 1.7% | 0.0% |
| Microsoft Outlook | 2 | 1.1% | 0.0% |
| TextInputHost | 2 | 1.1% | 0.0% |
| SnippingTool | 1 | 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 |
|---|---|---|
| Veeva Safety | Acrobat | 4 |
| Veeva Safety | Microsoft Teams | 4 |
| Microsoft Word | Veeva Safety | 3 |
| Acrobat | Veeva Safety | 3 |
| Microsoft Word | ONENOTE | 3 |
| ONENOTE | Microsoft Word | 3 |
| Microsoft Teams | Veeva Safety | 3 |
| Veeva Safety | Microsoft Excel | 3 |
| Veeva Safety | Microsoft Word | 3 |
| Microsoft Excel | Veeva Safety | 2 |
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 1,553 events 55 cases 23.8 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
145×
pyzeClick
<a>
"Show more"
75×
pyzeClick
<span>
"Generate Narrative from Outline"
65×
pyzeClick
<li>
"Narrative"
60×
pyzeClick
<span>
"Narrative"
48×
pyzeClick
<button>
"Create"
40×
pyzeClick
<span>
"Submissions & Distributions"
34×
pyzeClick
<li>
"Unknown"
33×
Phobos DOM 156 events 22 cases 4.6 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
21×
pyzeClick
<button>
"Advance"
15×
pyzeClick
<button>
"Save Comments"
13×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
9×
pyzeClick
<button>
"Send to process"
6×
inputld
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
5×
pyzeClick
<a>
"Cases"
5×
pyzeClick
<div>
"Case version number · First Inbox Item · Compound · Sel"
4×
Microsoft Word Document 54 events 20 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 525025 (English) (26253159_0_3_342205)
211×
file
V-Narrative 542512 (English) (26522671_0_2_344761)
189×
file
V-Narrative 531284 (English) (26355806_0_2_340584)
156×
file
V-Narrative 523170 (English) (26221379_0_2_340465)
144×
file
V-Narrative 433845 (English) (24804813_0_2_345392)
137×
file
V-Narrative 530932 (English) (26349278_0_2_339374)
118×
file
V-Narrative 537919 (English) (26452534_0_2_341686)
99×
file
V-Narrative 498632 (English) (25800349_0_2_340329)
87×
Microsoft Teams App 44 events 21 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
534 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Acrobat App 43 events 16 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
390 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 19 events 10 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
102×
file
All PV Smart Submissions 28-NOV-2025 PM
12×
file
records_as_of_2025_12_01_EST
12×
file
Case tracking list
9×
file
Priorización
5×
Microsoft Outlook Message 9 events 7 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
ONENOTE Document 9 events 5 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
gpteal.merck.com DOM 6 events 3 cases
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
olk Message 6 events 4 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.10
Swivel Rate
events in non-primary app / total
0.98
Zero-Edit Visit Rate
inspection-only clicks
1.19
Avg Revisit Depth
max repeats of any activity
0.00
Cross-App Time Share
time outside primary app
Meta Score Distribution
Instance count in each 10-point adherence bucket. Right-skewed = healthy execution; long left tail = concentrated pain points.
0
10
20
30
40
50
60
70
80
90
Top Concern Vector
Zero-Edit Visit Rate — current value 0.98
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 0.98
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.19
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.10
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 |
|---|---|---|---|---|---|---|---|
| 2351809 | Analyst 01 | 48.9 | 16.8m | 14 | 0.79 | 0.00 | 5 |
| 2357099 | Analyst 12 | 56.0 | 1.8m | 5 | 0.80 | 0.00 | 2 |
| 2347303 | Analyst 02 | 62.5 | 24.8m | 8 | 0.50 | 0.00 | 3 |
| 2352746 | Analyst 02 | 65.5 | 12.7m | 10 | 0.40 | 0.00 | 3 |
| 2354665 | Analyst 02 | 67.5 | 0.5m | 2 | 0.50 | 0.00 | 1 |
Users on This Task
9 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) →
05
04
15
14
13
12
09
01
02
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 05 | 4 | 2.0 | 85.6 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 05's Dominant Sequence
— 2 of 4 instances
1
Generate Narrative from Outline
This analyst exhibits 3 distinct sequences across their 4 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 04 | 15 | 2.1 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 10 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 04's Dominant Sequence
— 10 of 15 instances
1
Generate Narrative from Outline
This analyst exhibits 6 distinct sequences across their 15 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 15 | 3 | 3.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 15's Dominant Sequence
— 1 of 3 instances
1
Generate Narrative from Outline
2
Select Case Assessment Results
· reference: Open Details Section
This analyst exhibits 3 distinct sequences across their 3 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 14 | 5 | 2.5 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 5 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 14's Dominant Sequence
— 3 of 5 instances
1
Generate Narrative from Outline
This analyst exhibits 3 distinct sequences across their 5 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 1 | 1.0 | 82.5 | Zero-Edit Visits | 0.00 | 0.00 | 0 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 13's Dominant Sequence
— 1 of 1 instances
1
Generate Narrative from Outline
|
||||||||
| ▸ | Analyst 12 | 8 | 1.6 | 79.8 | Zero-Edit Visits | 0.10 | 0.00 | 10 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 12's Dominant Sequence
— 4 of 8 instances
1
Generate Narrative from Outline
This analyst exhibits 5 distinct sequences across their 8 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 7 | 1.8 | 78.3 | Zero-Edit Visits | 0.13 | 0.00 | 24 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 09's Dominant Sequence
— 2 of 7 instances
1
Generate Narrative from Outline
This analyst exhibits 6 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 13 | 1.9 | 77.0 | Zero-Edit Visits | 0.16 | 0.00 | 23 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 01's Dominant Sequence
— 8 of 13 instances
1
Generate Narrative from Outline
This analyst exhibits 6 distinct sequences across their 13 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 7 | 1.2 | 70.4 | Zero-Edit Visits | 0.34 | 0.00 | 56 |
|
Reference SoP
1
Generate Narrative from Outline
2
Open Details Section
3
Navigate to Cases
4
gpteal.merck.com
5
Microsoft Word
6
Select Patient
7
Microsoft Word
Analyst 02's Dominant Sequence
— 1 of 7 instances
1
Generate Narrative from Outline
2
View Site Narrative
· reference: Open Details Section
3
View Case Description
· reference: Navigate to Cases
4
Show More Details
· reference: gpteal.merck.com
5
View Investigator Information
· reference: Microsoft Word
6
SnippingTool
· reference: Select Patient
7
Show More Details
· reference: Microsoft Word
8
Microsoft PowerPoint
This analyst exhibits 7 distinct sequences across their 7 instances; the dominant one is shown.
|
||||||||