Task-Boundary Mode
How task-instance boundaries are drawn from the event stream. Applies to every Task SoP, Step SoP, and Variants view.
Assessment
📝 Task SoP
Mined from 702 instances
Categorize Case
How 8 analysts execute this task across 297 cases — auto-generated Standard Operating Procedure with adherence scoring.
Sub-Task Filter
Reshapes all tabs below to just this anchor within the composite task.
90.1
Avg Adherence Score (0-100)
Median 97.5 · Range 45.2–97.5
17.9 min
Avg Task Duration
0.0 min median
1.2
Avg Sub-Steps per Instance
10.4% have multiple sub-steps
1.1%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
1.1% of instances follow this exactly
Categorize Case — Mined SoP
The most common sub-step sequence observed across 742 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Enter Case Type
2.3s
Phobos
54%
2
Navigate to Cases
0.0s
Veeva Safety
1%
Variant Explorer
Reference
Categorize Case Reference Path
8 instances · 3 analysts
This is the mined reference path — 8 instances across 3 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Categorize Case.
Genuine Complexity
Complex-Case Execution Path
344 instances · 8 analysts
· Δ adherence +1.2
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
×
Navigate to Cases
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 91.4 vs reference 90.1 (delta +1.2). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
181 instances · 7 analysts
· Δ adherence +4.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
×
Navigate to Cases
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 94.8 vs reference 90.1 (delta +4.7). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
140 instances · 7 analysts
· Δ adherence +1.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter MARRS Case ID
reference at this step: Enter Case Type
×
Navigate to Cases
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 91.7 vs reference 90.1 (delta +1.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Veeva Safety
4 instances · 3 analysts
· Δ adherence -18.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Start Approval Workflow
reference at this step: Navigate to Cases
Cross-App Swivel — 3 analysts pivot into Veeva Safety during this task (0% of task time in non-primary apps, 50% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Genuine Complexity
Complex-Case Execution Path
3 instances · 3 analysts
· Δ adherence -0.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter MARRS Case ID
reference at this step: Enter Case Type
2
Enter Inbox ID
reference at this step: Navigate to Cases
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 90.0 vs reference 90.1 (delta -0.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Microsoft Teams
3 instances · 3 analysts
· Δ adherence -17.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Microsoft Teams
reference at this step: Navigate to Cases
Cross-App Swivel — 3 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 Veeva Safety
3 instances · 1 analysts
· Δ adherence -22.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Navigate to Home
reference at this step: Navigate to Cases
Cross-App Swivel — 1 analysts pivot into Veeva Safety 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 Teams
2 instances · 1 analysts
· Δ adherence -15.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter MARRS Case ID
reference at this step: Enter Case Type
2
Microsoft Teams
reference at this step: Navigate to Cases
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 Veeva Safety
2 instances · 2 analysts
· Δ adherence -15.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
2
View Case
reference at this step: Navigate to Cases
Cross-App Swivel — 2 analysts pivot into Veeva Safety 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 Veeva Safety
2 instances · 2 analysts
· Δ adherence -15.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
View Validation Results
reference at this step: Navigate to Cases
Cross-App Swivel — 2 analysts pivot into Veeva Safety during this task (0% of task time in non-primary apps, 50% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Cross-App Swivel
Cross-App Lookup to Microsoft Excel
2 instances · 2 analysts
· Δ adherence -22.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Microsoft Excel
reference at this step: Navigate to Cases
Cross-App Swivel — 2 analysts pivot into Microsoft Excel during this task (0% of task time in non-primary apps, 50% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Cross-App Swivel
Cross-App Lookup to Veeva Safety
2 instances · 2 analysts
· Δ adherence -25.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Mark Data Entry Complete
reference at this step: Navigate to Cases
Cross-App Swivel — 2 analysts pivot into Veeva Safety during this task (35% of task time in non-primary apps, 50% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 748 | 83.2% | 95.8% |
| Veeva Safety | 118 | 13.1% | 2.5% |
| Microsoft Teams | 8 | 0.9% | 0.0% |
| Acrobat | 7 | 0.8% | 0.0% |
| collaboration.merck.com | 5 | 0.6% | 0.0% |
| Microsoft Word | 3 | 0.3% | 0.0% |
| Microsoft Excel | 3 | 0.3% | 0.0% |
| Quality Vault | 2 | 0.2% | 1.7% |
| usc-excel.officeapps.live.com | 1 | 0.1% | 0.0% |
| ONENOTE | 1 | 0.1% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Phobos | Veeva Safety | 51 |
| Phobos | Microsoft Teams | 5 |
| Microsoft Word | Acrobat | 3 |
| Veeva Safety | Phobos | 3 |
| Phobos | collaboration.merck.com | 3 |
| Acrobat | Veeva Safety | 3 |
| Veeva Safety | Microsoft Word | 2 |
| Phobos | Acrobat | 2 |
| Veeva Safety | Microsoft Teams | 2 |
| Phobos | Quality Vault | 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.
Phobos DOM 4,209 events 297 cases 109.1 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
462×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
451×
pyzeClick
<button>
"Save Comments"
328×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
310×
pyzeClick
<button>
"Advance"
216×
pyzeClick
<button>
"Send to process"
208×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
152×
pyzeClick
<a>
"Cases"
145×
Veeva Safety DOM 3,897 events 262 cases 47.7 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
539×
pyzeClick
<a>
"Show more"
180×
pyzeClick
<span>
"Adverse Events"
125×
pyzeClick
<span>
"Narrative"
116×
pyzeClick
<li>
"Narrative"
105×
pyzeClick
<li>
"Details"
90×
pyzeClick
<span>
"Approval Complete"
85×
pyzeClick
<span>
"Products"
82×
Microsoft Word Document 101 events 46 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 538518 (English) (26459147_0_1_344628)
198×
file
V-Narrative 542172 (English) (26517690_0_2_345412)
181×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
file
V-Narrative 538589 (English) (26461195_0_1_346748)
160×
file
V-Narrative 527377 (English) (26286859_0_1_339572)
113×
file
V-Narrative 525025 (English) (26253159_0_3_342205)
109×
file
V-Narrative 537919 (English) (26452534_0_2_341686)
99×
file
V-Narrative 541519 (English) (26506626_0_2_342362)
75×
Acrobat App 99 events 36 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,357 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 90 events 55 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,141 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
collaboration.merck.com App 84 events 40 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
810 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 71 events 34 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Outlook Message 56 events 31 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
RIM Vault DOM 39 events 5 cases 0.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
usc-excel.officeapps.live.com Document 32 events 7 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.05
Swivel Rate
events in non-primary app / total
0.38
Zero-Edit Visit Rate
inspection-only clicks
1.01
Avg Revisit Depth
max repeats of any activity
0.01
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 0.38
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 0.38
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.01
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.05
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2353407 | Analyst 09 | 45.2 | 12.0m | 11 | 0.91 | 0.00 | 5 |
| 2353221 | Analyst 13 | 46.9 | 22.4m | 16 | 0.94 | 0.00 | 4 |
| 2264485 | Analyst 01 | 50.0 | 8.0m | 10 | 0.90 | 0.23 | 1 |
| 2354095 | Analyst 02 | 56.0 | 0.9m | 5 | 0.80 | 0.00 | 2 |
| 2351809 | Analyst 01 | 60.0 | 4.5m | 4 | 0.75 | 0.00 | 1 |
Users on This Task
8 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
10
03
05
12
09
01
13
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 10 | 23 | 2.6 | 93.6 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 10's Dominant Sequence
— 23 of 23 instances
1
Enter Case Type
|
||||||||
| ▸ | Analyst 03 | 156 | 12.0 | 92.3 | Zero-Edit Visits | 0.03 | 0.00 | 60 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 03's Dominant Sequence
— 76 of 156 instances
1
Enter Case Type
This analyst exhibits 13 distinct sequences across their 156 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 59 | 6.6 | 90.7 | Zero-Edit Visits | 0.04 | 0.00 | 10,067 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 05's Dominant Sequence
— 24 of 59 instances
1
Enter Case Type
This analyst exhibits 7 distinct sequences across their 59 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 104 | 11.6 | 90.7 | Zero-Edit Visits | 0.04 | 0.01 | 14 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 12's Dominant Sequence
— 41 of 104 instances
1
Enter Case Type
This analyst exhibits 12 distinct sequences across their 104 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 106 | 7.1 | 90.2 | Zero-Edit Visits | 0.03 | 0.00 | 20 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 09's Dominant Sequence
— 66 of 106 instances
1
Enter Case Type
This analyst exhibits 7 distinct sequences across their 106 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 128 | 9.8 | 89.2 | Zero-Edit Visits | 0.06 | 0.02 | 39 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 01's Dominant Sequence
— 52 of 128 instances
1
Enter Case Type
This analyst exhibits 16 distinct sequences across their 128 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 98 | 8.9 | 88.6 | Zero-Edit Visits | 0.07 | 0.00 | 90 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 13's Dominant Sequence
— 37 of 98 instances
1
Enter Case Type
This analyst exhibits 16 distinct sequences across their 98 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 68 | 8.5 | 86.6 | Zero-Edit Visits | 0.12 | 0.00 | 2,987 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 02's Dominant Sequence
— 25 of 68 instances
1
Enter Case Type
This analyst exhibits 17 distinct sequences across their 68 instances; the dominant one is shown.
|
||||||||
88.0
Avg Adherence Score (0-100)
Median 97.5 · Range 45.2–97.5
25.5 min
Avg Task Duration
0.0 min median
1.3
Avg Sub-Steps per Instance
14.4% have multiple sub-steps
—
Reference Path Coverage
402 sub-task instances
Reference Baseline
Categorize Case — Mined SoP
The most common sub-step sequence observed across 402 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Enter Case Type
2
Navigate to Cases
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 402 | 75.6% | 92.8% |
| Veeva Safety | 101 | 19.0% | 3.5% |
| Acrobat | 7 | 1.3% | 0.0% |
| collaboration.merck.com | 5 | 0.9% | 0.0% |
| Microsoft Teams | 5 | 0.9% | 0.0% |
| Microsoft Word | 3 | 0.6% | 0.0% |
| Microsoft Excel | 3 | 0.6% | 0.0% |
| Quality Vault | 2 | 0.4% | 3.8% |
| usc-excel.officeapps.live.com | 1 | 0.2% | 0.0% |
| ONENOTE | 1 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Phobos | Veeva Safety | 44 |
| Acrobat | Veeva Safety | 3 |
| Microsoft Word | Acrobat | 3 |
| Phobos | collaboration.merck.com | 3 |
| Veeva Safety | Microsoft Word | 2 |
| Phobos | Microsoft Teams | 2 |
| Phobos | Acrobat | 2 |
| Veeva Safety | Acrobat | 1 |
| collaboration.merck.com | Veeva Safety | 1 |
| Phobos | CelonisTaskMining.Client | 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 3,297 events 196 cases 42.0 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
488×
pyzeClick
<a>
"Show more"
159×
pyzeClick
<span>
"Narrative"
105×
pyzeClick
<span>
"Adverse Events"
99×
pyzeClick
<li>
"Narrative"
88×
pyzeClick
<span>
"Approval Complete"
83×
pyzeClick
<span>
"Start Approval Workflow"
71×
pyzeClick
<li>
"Details"
69×
Phobos DOM 3,176 events 229 cases 83.6 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
291×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
181×
pyzeClick
<button>
"Send to process"
171×
pyzeClick
<button>
"Save Comments"
168×
pyzeClick
<button>
"Advance"
159×
pyzeClick
<a>
"Cases"
107×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
105×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
74×
Microsoft Word Document 96 events 41 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 538518 (English) (26459147_0_1_344628)
198×
file
V-Narrative 542172 (English) (26517690_0_2_345412)
181×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
file
V-Narrative 538589 (English) (26461195_0_1_346748)
160×
file
V-Narrative 527377 (English) (26286859_0_1_339572)
113×
file
V-Narrative 525025 (English) (26253159_0_3_342205)
109×
file
V-Narrative 537919 (English) (26452534_0_2_341686)
99×
file
V-Narrative 541519 (English) (26506626_0_2_342362)
75×
Acrobat App 82 events 30 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,254 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 57 events 35 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
791 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Outlook Message 51 events 27 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
subj
Incorrect article uploaded in CLIC 940845
37×
subj
<Due today><Void request><GPVCM POLAND> Case # 2326062
23×
subj
<compensación> Case#2351734
17×
subj
Due today reports on 01-DEC-2025
13×
subj
Untitled
8×
subj
RE: <compensación> Case#2309077
8×
subj
<OT 01-DEC PLEASE REVIEW> Organizacion OT
7×
subj
<Due today><Void request><GPVCM POLAND> Case #
4×
collaboration.merck.com App 50 events 29 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Microsoft Excel Document 47 events 23 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
usc-excel.officeapps.live.com Document 16 events 3 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
ONENOTE Document 11 events 6 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.07
Swivel Rate
events in non-primary app / total
0.46
Zero-Edit Visit Rate
inspection-only clicks
1.03
Avg Revisit Depth
max repeats of any activity
0.01
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 0.46
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 0.46
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.03
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.08
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2353407 | Analyst 09 | 45.2 | 12.0m | 11 | 0.91 | 0.00 | 5 |
| 2353221 | Analyst 13 | 46.9 | 22.4m | 16 | 0.94 | 0.00 | 4 |
| 2264485 | Analyst 01 | 50.0 | 8.0m | 10 | 0.90 | 0.23 | 1 |
| 2354095 | Analyst 02 | 56.0 | 0.9m | 5 | 0.80 | 0.00 | 2 |
| 2351809 | Analyst 01 | 60.0 | 4.5m | 4 | 0.75 | 0.00 | 1 |
Users on This Task
8 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
10
03
05
09
01
12
13
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 10 | 23 | 2.6 | 93.6 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 10's Dominant Sequence
— 23 of 23 instances
1
Enter Case Type
|
||||||||
| ▸ | Analyst 03 | 86 | 6.6 | 90.0 | Zero-Edit Visits | 0.06 | 0.01 | 51 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 03's Dominant Sequence
— 76 of 86 instances
1
Enter Case Type
This analyst exhibits 10 distinct sequences across their 86 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 29 | 3.6 | 88.3 | Zero-Edit Visits | 0.09 | 0.01 | 10,057 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 05's Dominant Sequence
— 24 of 29 instances
1
Enter Case Type
This analyst exhibits 4 distinct sequences across their 29 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 71 | 4.7 | 87.8 | Zero-Edit Visits | 0.04 | 0.00 | 16 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 09's Dominant Sequence
— 66 of 71 instances
1
Enter Case Type
This analyst exhibits 4 distinct sequences across their 71 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 62 | 4.8 | 87.7 | Zero-Edit Visits | 0.09 | 0.02 | 26 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 01's Dominant Sequence
— 52 of 62 instances
1
Enter Case Type
This analyst exhibits 11 distinct sequences across their 62 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 51 | 5.7 | 86.6 | Zero-Edit Visits | 0.08 | 0.01 | 10 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 12's Dominant Sequence
— 41 of 51 instances
1
Enter Case Type
This analyst exhibits 8 distinct sequences across their 51 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 48 | 4.8 | 86.2 | Zero-Edit Visits | 0.13 | 0.01 | 76 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 13's Dominant Sequence
— 37 of 48 instances
1
Enter Case Type
This analyst exhibits 12 distinct sequences across their 48 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 32 | 5.3 | 84.4 | Zero-Edit Visits | 0.12 | 0.01 | 14 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 02's Dominant Sequence
— 25 of 32 instances
1
Enter Case Type
This analyst exhibits 8 distinct sequences across their 32 instances; the dominant one is shown.
|
||||||||
90.8
Avg Adherence Score (0-100)
Median 97.5 · Range 61.5–97.5
19.9 min
Avg Task Duration
0.1 min median
1.1
Avg Sub-Steps per Instance
6.7% have multiple sub-steps
—
Reference Path Coverage
150 sub-task instances
Reference Baseline
Categorize Case — Mined SoP
The most common sub-step sequence observed across 150 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Enter MARRS Case ID
2
Microsoft Teams
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 154 | 93.3% | 100.0% |
| Veeva Safety | 7 | 4.2% | 0.0% |
| Microsoft Teams | 3 | 1.8% | 0.0% |
| Notepad | 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 |
|---|---|---|
| Phobos | Veeva Safety | 3 |
| Veeva Safety | Microsoft Teams | 1 |
| Phobos | Microsoft Teams | 1 |
| Phobos | Notepad | 1 |
| Microsoft Teams | Phobos | 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.
Phobos DOM 1,573 events 114 cases 41.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
371×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
264×
pyzeClick
<button>
"Proceed"
212×
pyzeClick
<button>
"Save Comments"
190×
pyzeClick
<button>
"Push to Case Processing"
117×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
75×
pyzeClick
<div>
"Edit info and/or add comments · × · Compound · Select C"
74×
pyzeClick
<button>
"Advance"
64×
Veeva Safety DOM 585 events 96 cases 3.7 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Adverse Events"
38×
pyzeClick
<a>
"Show more"
27×
pyzeClick
<span>
"This Case is Locked"
26×
pyzeClick
<span>
"Products"
19×
pyzeClick
<button>
"Save"
18×
pyzeClick
<li>
"This Case is Locked"
18×
pyzeClick
<li>
"Narrative"
17×
pyzeClick
<li>
"Details"
17×
collaboration.merck.com App 34 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
360 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 24 events 13 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Priorización
186×
file
Daily tracking list (PV-SMART) copy
56×
file
MERCK2519807_ARGPRO_20251126220317
55×
file
MERCK2518114_ARGPRO_20251125190215 (1)
18×
file
Query 21-NOV-2025
9×
file
210-PV007-006 Important Medical Event (IME) List
2×
Microsoft Teams App 23 events 18 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
282 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.
RIM Vault DOM 20 events 4 cases 0.1 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Continue"
2×
pyzeClick
<div>
"Products"
2×
pyzeClick
<a>
"Consortium Agreement NCI MK-3475"
2×
pyzeClick
<div>
"BDL Agreement to Clinical Study > Clinical Study"
2×
pyzeClick
<span>
"Agreement Documents"
2×
pyzeClick
<div>
"This report requires you to complete the report v"
2×
pyzeClick
<span>
"Back to previous page"
1×
pyzeClick
<div>
"Show in Tab"
1×
Acrobat App 16 events 6 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
usc-excel.officeapps.live.com Document 16 events 4 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Quality Vault DOM 7 events 5 cases 5.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Microsoft Outlook Message 6 events 5 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.
0.02
Swivel Rate
events in non-primary app / total
0.40
Zero-Edit Visit Rate
inspection-only clicks
1.0
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 0.40
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 0.40
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.00
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.02
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 |
|---|---|---|---|---|---|---|---|
| 2291017 | Analyst 02 | 61.5 | 5.5m | 5 | 0.80 | 0.00 | 1 |
| 2219900 | Analyst 02 | 63.7 | 1.9m | 4 | 0.75 | 0.00 | 1 |
| 2219403 | Analyst 02 | 75.0 | 2955.9m | 2 | 0.50 | 0.00 | 1 |
| 2341212 | Analyst 02 | 75.0 | 2.0m | 2 | 0.50 | 0.00 | 1 |
| 2219701 | Analyst 02 | 75.0 | 1.4m | 2 | 0.50 | 0.00 | 1 |
Users on This Task
7 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) →
09
03
12
01
05
02
13
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 09 | 14 | 3.5 | 97.0 | Revisit Depth | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 09's Dominant Sequence
— 13 of 14 instances
1
Enter MARRS Case ID
This analyst exhibits 2 distinct sequences across their 14 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 24 | 2.2 | 94.4 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 03's Dominant Sequence
— 24 of 24 instances
1
Enter MARRS Case ID
|
||||||||
| ▸ | Analyst 12 | 27 | 3.9 | 93.3 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 12's Dominant Sequence
— 25 of 27 instances
1
Enter MARRS Case ID
This analyst exhibits 3 distinct sequences across their 27 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 29 | 4.1 | 89.7 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 01's Dominant Sequence
— 29 of 29 instances
1
Enter MARRS Case ID
|
||||||||
| ▸ | Analyst 05 | 14 | 3.5 | 88.9 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 05's Dominant Sequence
— 14 of 14 instances
1
Enter MARRS Case ID
|
||||||||
| ▸ | Analyst 02 | 17 | 2.8 | 86.8 | Zero-Edit Visits | 0.21 | 0.00 | 2,973 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 02's Dominant Sequence
— 10 of 17 instances
1
Enter MARRS Case ID
This analyst exhibits 7 distinct sequences across their 17 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 25 | 3.1 | 86.1 | Zero-Edit Visits | 0.00 | 0.00 | 8 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 13's Dominant Sequence
— 25 of 25 instances
1
Enter MARRS Case ID
|
||||||||
94.1
Avg Adherence Score (0-100)
Median 97.5 · Range 71.3–97.5
0.2 min
Avg Task Duration
0.1 min median
1.1
Avg Sub-Steps per Instance
4.7% have multiple sub-steps
—
Reference Path Coverage
190 sub-task instances
Reference Baseline
Categorize Case — Mined SoP
The most common sub-step sequence observed across 190 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Select Report Type
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 192 | 95.0% | 97.2% |
| Veeva Safety | 10 | 5.0% | 2.8% |
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 |
|---|---|---|
| Phobos | Veeva Safety | 7 |
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.
Phobos DOM 2,100 events 143 cases 56.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
403×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
287×
pyzeClick
<button>
"Proceed"
256×
pyzeClick
<button>
"Save Comments"
241×
pyzeClick
<button>
"Push to Case Processing"
118×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
115×
pyzeClick
<button>
"Advance"
103×
pyzeClick
<div>
"Edit info and/or add comments · × · Compound · Select C"
77×
Veeva Safety DOM 1,092 events 128 cases 10.9 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
111×
pyzeClick
<span>
"Adverse Events"
48×
pyzeClick
<a>
"Show more"
47×
pyzeClick
<span>
"Products"
35×
pyzeClick
<li>
"Details"
34×
pyzeClick
<span>
"Narrative"
29×
pyzeClick
<span>
"Case Assessment Results"
26×
pyzeClick
<span>
"This Case is Locked"
26×
RIM Vault DOM 39 events 5 cases 0.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Continue"
4×
pyzeClick
<span>
"Agreement Documents"
3×
pyzeClick
<a>
"MK-2870"
2×
pyzeClick
<div>
"Products"
2×
pyzeClick
<a>
"Consortium Agreement NCI MK-3475"
2×
pyzeClick
<a>
"Kelun"
2×
pyzeClick
<div>
"BDL Agreement to Clinical Study > Clinical Study"
2×
pyzeClick
<div>
"This report requires you to complete the report v"
2×
Microsoft Teams App 34 events 24 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
399 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
collaboration.merck.com App 34 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
323 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 26 events 14 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Priorización
182×
file
Daily tracking list (PV-SMART) copy
72×
file
MERCK2519807_ARGPRO_20251126220317
55×
file
MERCK2518114_ARGPRO_20251125190215 (1)
18×
file
Query 21-NOV-2025
9×
file
Book1
3×
file
210-PV007-006 Important Medical Event (IME) List
2×
Acrobat App 20 events 7 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
usc-excel.officeapps.live.com Document 16 events 4 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Word Document 9 events 7 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Quality Vault DOM 8 events 5 cases 5.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.02
Swivel Rate
events in non-primary app / total
0.19
Zero-Edit Visit Rate
inspection-only clicks
1.0
Avg Revisit Depth
max repeats of any activity
0.01
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 0.19
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 0.19
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.00
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.02
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2353955 | Analyst 01 | 71.3 | 0.2m | 3 | 0.33 | 0.37 | 1 |
| 2353954 | Analyst 01 | 72.2 | 0.3m | 3 | 0.33 | 0.34 | 1 |
| 2347125 | Analyst 01 | 74.2 | 6.3m | 3 | 0.33 | 0.28 | 1 |
| 2275847 | Analyst 01 | 75.0 | 0.0m | 2 | 0.50 | 0.00 | 1 |
| 2346744 | Analyst 13 | 75.0 | 0.1m | 2 | 0.50 | 0.00 | 1 |
Users on This Task
7 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
12
13
03
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 | 16 | 4.0 | 96.6 | Revisit Depth | 0.00 | 0.00 | 9 |
|
Reference SoP
1
Select Report Type
Analyst 05's Dominant Sequence
— 15 of 16 instances
1
Select Report Type
This analyst exhibits 2 distinct sequences across their 16 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 26 | 3.7 | 95.8 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Select Report Type
Analyst 12's Dominant Sequence
— 26 of 26 instances
1
Select Report Type
|
||||||||
| ▸ | Analyst 13 | 25 | 3.6 | 95.7 | Revisit Depth | 0.02 | 0.00 | 6 |
|
Reference SoP
1
Select Report Type
Analyst 13's Dominant Sequence
— 23 of 25 instances
1
Select Report Type
This analyst exhibits 3 distinct sequences across their 25 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 46 | 3.5 | 95.4 | Zero-Edit Visits | 0.01 | 0.00 | 7 |
|
Reference SoP
1
Select Report Type
Analyst 03's Dominant Sequence
— 45 of 46 instances
1
Select Report Type
This analyst exhibits 2 distinct sequences across their 46 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 21 | 4.2 | 93.9 | Zero-Edit Visits | 0.00 | 0.00 | 3 |
|
Reference SoP
1
Select Report Type
Analyst 09's Dominant Sequence
— 21 of 21 instances
1
Select Report Type
|
||||||||
| ▸ | Analyst 01 | 37 | 4.6 | 91.2 | Zero-Edit Visits | 0.04 | 0.03 | 9 |
|
Reference SoP
1
Select Report Type
Analyst 01's Dominant Sequence
— 33 of 37 instances
1
Select Report Type
This analyst exhibits 4 distinct sequences across their 37 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 19 | 2.7 | 90.0 | Zero-Edit Visits | 0.03 | 0.00 | 1 |
|
Reference SoP
1
Select Report Type
Analyst 02's Dominant Sequence
— 18 of 19 instances
1
Select Report Type
This analyst exhibits 2 distinct sequences across their 19 instances; the dominant one is shown.
|
||||||||
90.1
Avg Adherence Score (0-100)
Median 97.5 · Range 45.2–97.5
18.0 min
Avg Task Duration
0.0 min median
1.2
Avg Sub-Steps per Instance
10.6% have multiple sub-steps
1.1%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
1.1% of instances follow this exactly
Categorize Case — Mined SoP
The most common sub-step sequence observed across 742 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Enter Case Type
2.3s
Phobos
54%
2
Navigate to Cases
0.0s
Veeva Safety
1%
Variant Explorer
Reference
Categorize Case Reference Path
8 instances · 3 analysts
This is the mined reference path — 8 instances across 3 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Categorize Case.
Genuine Complexity
Complex-Case Execution Path
342 instances · 8 analysts
· Δ adherence +1.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
×
Navigate to Cases
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 91.4 vs reference 90.1 (delta +1.3). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
181 instances · 7 analysts
· Δ adherence +4.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
×
Navigate to Cases
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 94.8 vs reference 90.1 (delta +4.8). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
140 instances · 7 analysts
· Δ adherence +1.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter MARRS Case ID
reference at this step: Enter Case Type
×
Navigate to Cases
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 91.7 vs reference 90.1 (delta +1.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Veeva Safety
4 instances · 3 analysts
· Δ adherence -18.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Start Approval Workflow
reference at this step: Navigate to Cases
Cross-App Swivel — 3 analysts pivot into Veeva Safety during this task (0% of task time in non-primary apps, 50% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Genuine Complexity
Complex-Case Execution Path
3 instances · 3 analysts
· Δ adherence -0.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter MARRS Case ID
reference at this step: Enter Case Type
2
Enter Inbox ID
reference at this step: Navigate to Cases
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 90.0 vs reference 90.1 (delta -0.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Veeva Safety
3 instances · 2 analysts
· Δ adherence -15.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
View Validation Results
reference at this step: Navigate to Cases
Cross-App Swivel — 2 analysts pivot into Veeva Safety 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 Veeva Safety
3 instances · 1 analysts
· Δ adherence -22.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Navigate to Home
reference at this step: Navigate to Cases
Cross-App Swivel — 1 analysts pivot into Veeva Safety 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 Veeva Safety
2 instances · 2 analysts
· Δ adherence -15.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
2
View Case
reference at this step: Navigate to Cases
Cross-App Swivel — 2 analysts pivot into Veeva Safety 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 Teams
2 instances · 2 analysts
· Δ adherence -18.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Microsoft Teams
reference at this step: Navigate to Cases
Cross-App Swivel — 2 analysts pivot into Microsoft Teams during this task (0% of task time in non-primary apps, 50% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Cross-App Swivel
Cross-App Lookup to Phobos
2 instances · 1 analysts
· Δ adherence -18.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
2
Select List Item
reference at this step: Navigate to Cases
3
Start Action
extra step (reference ends at step 2)
Cross-App Swivel — 1 analysts pivot into Phobos during this task (36% of task time in non-primary apps, 33% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Cross-App Swivel
Cross-App Lookup to Microsoft Teams
1 instances · 1 analysts
· Δ adherence -15.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter MARRS Case ID
reference at this step: Enter Case Type
2
Microsoft Teams
reference at this step: Navigate to Cases
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 Veeva Safety
1 instances · 1 analysts
· Δ adherence -15.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter MARRS Case ID
reference at this step: Enter Case Type
2
Select Contacts
reference at this step: Navigate to Cases
Cross-App Swivel — 1 analysts pivot into Veeva Safety during this task (0% of task time in non-primary apps, 50% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 748 | 83.0% | 95.8% |
| Veeva Safety | 120 | 13.3% | 2.5% |
| Microsoft Teams | 8 | 0.9% | 0.0% |
| Acrobat | 7 | 0.8% | 0.0% |
| collaboration.merck.com | 5 | 0.6% | 0.0% |
| Microsoft Word | 3 | 0.3% | 0.0% |
| Microsoft Excel | 3 | 0.3% | 0.0% |
| Quality Vault | 2 | 0.2% | 1.7% |
| Notepad | 1 | 0.1% | 0.0% |
| ONENOTE | 1 | 0.1% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Phobos | Veeva Safety | 54 |
| Phobos | Microsoft Teams | 5 |
| Phobos | collaboration.merck.com | 3 |
| Acrobat | Veeva Safety | 3 |
| Veeva Safety | Phobos | 3 |
| Microsoft Word | Acrobat | 3 |
| Veeva Safety | Microsoft Word | 2 |
| Phobos | Acrobat | 2 |
| Veeva Safety | Microsoft Teams | 2 |
| Microsoft Excel | Phobos | 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.
Phobos DOM 4,209 events 297 cases 109.1 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
462×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
451×
pyzeClick
<button>
"Save Comments"
328×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
310×
pyzeClick
<button>
"Advance"
216×
pyzeClick
<button>
"Send to process"
208×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
152×
pyzeClick
<a>
"Cases"
145×
Veeva Safety DOM 3,897 events 262 cases 47.7 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
539×
pyzeClick
<a>
"Show more"
180×
pyzeClick
<span>
"Adverse Events"
125×
pyzeClick
<span>
"Narrative"
116×
pyzeClick
<li>
"Narrative"
105×
pyzeClick
<li>
"Details"
90×
pyzeClick
<span>
"Approval Complete"
85×
pyzeClick
<span>
"Products"
82×
Microsoft Word Document 101 events 46 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 538518 (English) (26459147_0_1_344628)
198×
file
V-Narrative 542172 (English) (26517690_0_2_345412)
181×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
file
V-Narrative 538589 (English) (26461195_0_1_346748)
160×
file
V-Narrative 527377 (English) (26286859_0_1_339572)
113×
file
V-Narrative 525025 (English) (26253159_0_3_342205)
109×
file
V-Narrative 537919 (English) (26452534_0_2_341686)
99×
file
V-Narrative 541519 (English) (26506626_0_2_342362)
75×
Acrobat App 99 events 36 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,357 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 90 events 55 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,141 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
collaboration.merck.com App 84 events 40 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
810 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 71 events 34 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Outlook Message 56 events 31 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
RIM Vault DOM 39 events 5 cases 0.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
usc-excel.officeapps.live.com Document 32 events 7 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.05
Swivel Rate
events in non-primary app / total
0.38
Zero-Edit Visit Rate
inspection-only clicks
1.01
Avg Revisit Depth
max repeats of any activity
0.01
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 0.38
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 0.38
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.01
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.05
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2353407 | Analyst 09 | 45.2 | 12.0m | 11 | 0.91 | 0.00 | 5 |
| 2353221 | Analyst 13 | 46.9 | 22.4m | 16 | 0.94 | 0.00 | 4 |
| 2264485 | Analyst 01 | 50.0 | 8.0m | 10 | 0.90 | 0.23 | 1 |
| 2354095 | Analyst 02 | 56.0 | 0.9m | 5 | 0.80 | 0.00 | 2 |
| 2351809 | Analyst 01 | 60.0 | 4.5m | 4 | 0.75 | 0.00 | 1 |
Users on This Task
8 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
10
03
05
12
09
01
13
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 10 | 23 | 2.6 | 93.6 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 10's Dominant Sequence
— 23 of 23 instances
1
Enter Case Type
|
||||||||
| ▸ | Analyst 03 | 156 | 12.0 | 92.1 | Zero-Edit Visits | 0.04 | 0.00 | 70 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 03's Dominant Sequence
— 75 of 156 instances
1
Enter Case Type
This analyst exhibits 13 distinct sequences across their 156 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 59 | 6.6 | 90.7 | Zero-Edit Visits | 0.04 | 0.00 | 10,067 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 05's Dominant Sequence
— 24 of 59 instances
1
Enter Case Type
This analyst exhibits 7 distinct sequences across their 59 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 104 | 11.6 | 90.7 | Zero-Edit Visits | 0.04 | 0.01 | 14 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 12's Dominant Sequence
— 41 of 104 instances
1
Enter Case Type
This analyst exhibits 12 distinct sequences across their 104 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 106 | 7.1 | 90.2 | Zero-Edit Visits | 0.03 | 0.00 | 20 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 09's Dominant Sequence
— 66 of 106 instances
1
Enter Case Type
This analyst exhibits 7 distinct sequences across their 106 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 128 | 9.8 | 89.2 | Zero-Edit Visits | 0.06 | 0.02 | 39 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 01's Dominant Sequence
— 52 of 128 instances
1
Enter Case Type
This analyst exhibits 16 distinct sequences across their 128 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 98 | 8.9 | 88.4 | Zero-Edit Visits | 0.07 | 0.00 | 183 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 13's Dominant Sequence
— 36 of 98 instances
1
Enter Case Type
This analyst exhibits 17 distinct sequences across their 98 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 68 | 8.5 | 86.6 | Zero-Edit Visits | 0.12 | 0.00 | 2,987 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 02's Dominant Sequence
— 25 of 68 instances
1
Enter Case Type
This analyst exhibits 17 distinct sequences across their 68 instances; the dominant one is shown.
|
||||||||
87.9
Avg Adherence Score (0-100)
Median 97.5 · Range 45.2–97.5
25.8 min
Avg Task Duration
0.0 min median
1.3
Avg Sub-Steps per Instance
14.9% have multiple sub-steps
—
Reference Path Coverage
402 sub-task instances
Reference Baseline
Categorize Case — Mined SoP
The most common sub-step sequence observed across 402 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Enter Case Type
2
Navigate to Cases
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 402 | 75.3% | 92.8% |
| Veeva Safety | 103 | 19.3% | 3.5% |
| Acrobat | 7 | 1.3% | 0.0% |
| collaboration.merck.com | 5 | 0.9% | 0.0% |
| Microsoft Teams | 5 | 0.9% | 0.0% |
| Microsoft Word | 3 | 0.6% | 0.0% |
| Microsoft Excel | 3 | 0.6% | 0.0% |
| Quality Vault | 2 | 0.4% | 3.8% |
| usc-excel.officeapps.live.com | 1 | 0.2% | 0.0% |
| ONENOTE | 1 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Phobos | Veeva Safety | 44 |
| Phobos | collaboration.merck.com | 3 |
| Acrobat | Veeva Safety | 3 |
| Microsoft Word | Acrobat | 3 |
| Veeva Safety | Microsoft Word | 2 |
| Phobos | Acrobat | 2 |
| Phobos | Microsoft Teams | 2 |
| Veeva Safety | Phobos | 2 |
| Veeva Safety | Acrobat | 1 |
| Phobos | CelonisTaskMining.Client | 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 3,297 events 196 cases 42.0 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
488×
pyzeClick
<a>
"Show more"
159×
pyzeClick
<span>
"Narrative"
105×
pyzeClick
<span>
"Adverse Events"
99×
pyzeClick
<li>
"Narrative"
88×
pyzeClick
<span>
"Approval Complete"
83×
pyzeClick
<span>
"Start Approval Workflow"
71×
pyzeClick
<li>
"Details"
69×
Phobos DOM 3,176 events 229 cases 83.6 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
291×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
181×
pyzeClick
<button>
"Send to process"
171×
pyzeClick
<button>
"Save Comments"
168×
pyzeClick
<button>
"Advance"
159×
pyzeClick
<a>
"Cases"
107×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
105×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
74×
Microsoft Word Document 96 events 41 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 538518 (English) (26459147_0_1_344628)
198×
file
V-Narrative 542172 (English) (26517690_0_2_345412)
181×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
file
V-Narrative 538589 (English) (26461195_0_1_346748)
160×
file
V-Narrative 527377 (English) (26286859_0_1_339572)
113×
file
V-Narrative 525025 (English) (26253159_0_3_342205)
109×
file
V-Narrative 537919 (English) (26452534_0_2_341686)
99×
file
V-Narrative 541519 (English) (26506626_0_2_342362)
75×
Acrobat App 82 events 30 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,254 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 57 events 35 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
791 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Outlook Message 51 events 27 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
subj
Incorrect article uploaded in CLIC 940845
37×
subj
<Due today><Void request><GPVCM POLAND> Case # 2326062
23×
subj
<compensación> Case#2351734
17×
subj
Due today reports on 01-DEC-2025
13×
subj
Untitled
8×
subj
RE: <compensación> Case#2309077
8×
subj
<OT 01-DEC PLEASE REVIEW> Organizacion OT
7×
subj
<Due today><Void request><GPVCM POLAND> Case #
4×
collaboration.merck.com App 50 events 29 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Microsoft Excel Document 47 events 23 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
usc-excel.officeapps.live.com Document 16 events 3 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
ONENOTE Document 11 events 6 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.08
Swivel Rate
events in non-primary app / total
0.46
Zero-Edit Visit Rate
inspection-only clicks
1.03
Avg Revisit Depth
max repeats of any activity
0.01
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 0.46
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 0.46
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.03
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.08
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2353407 | Analyst 09 | 45.2 | 12.0m | 11 | 0.91 | 0.00 | 5 |
| 2353221 | Analyst 13 | 46.9 | 22.4m | 16 | 0.94 | 0.00 | 4 |
| 2264485 | Analyst 01 | 50.0 | 8.0m | 10 | 0.90 | 0.23 | 1 |
| 2354095 | Analyst 02 | 56.0 | 0.9m | 5 | 0.80 | 0.00 | 2 |
| 2351809 | Analyst 01 | 60.0 | 4.5m | 4 | 0.75 | 0.00 | 1 |
Users on This Task
8 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
10
03
05
09
01
12
13
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 10 | 23 | 2.6 | 93.6 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 10's Dominant Sequence
— 23 of 23 instances
1
Enter Case Type
|
||||||||
| ▸ | Analyst 03 | 86 | 6.6 | 89.8 | Zero-Edit Visits | 0.06 | 0.01 | 61 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 03's Dominant Sequence
— 75 of 86 instances
1
Enter Case Type
This analyst exhibits 10 distinct sequences across their 86 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 29 | 3.6 | 88.3 | Zero-Edit Visits | 0.09 | 0.01 | 10,057 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 05's Dominant Sequence
— 24 of 29 instances
1
Enter Case Type
This analyst exhibits 4 distinct sequences across their 29 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 71 | 4.7 | 87.8 | Zero-Edit Visits | 0.04 | 0.00 | 16 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 09's Dominant Sequence
— 66 of 71 instances
1
Enter Case Type
This analyst exhibits 4 distinct sequences across their 71 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 62 | 4.8 | 87.7 | Zero-Edit Visits | 0.09 | 0.02 | 26 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 01's Dominant Sequence
— 52 of 62 instances
1
Enter Case Type
This analyst exhibits 11 distinct sequences across their 62 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 51 | 5.7 | 86.6 | Zero-Edit Visits | 0.08 | 0.01 | 10 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 12's Dominant Sequence
— 41 of 51 instances
1
Enter Case Type
This analyst exhibits 8 distinct sequences across their 51 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 48 | 4.8 | 85.8 | Zero-Edit Visits | 0.14 | 0.01 | 169 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 13's Dominant Sequence
— 36 of 48 instances
1
Enter Case Type
This analyst exhibits 13 distinct sequences across their 48 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 32 | 5.3 | 84.4 | Zero-Edit Visits | 0.12 | 0.01 | 14 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 02's Dominant Sequence
— 25 of 32 instances
1
Enter Case Type
This analyst exhibits 8 distinct sequences across their 32 instances; the dominant one is shown.
|
||||||||
90.8
Avg Adherence Score (0-100)
Median 97.5 · Range 61.5–97.5
19.9 min
Avg Task Duration
0.1 min median
1.1
Avg Sub-Steps per Instance
6.7% have multiple sub-steps
—
Reference Path Coverage
150 sub-task instances
Reference Baseline
Categorize Case — Mined SoP
The most common sub-step sequence observed across 150 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Enter MARRS Case ID
2
Microsoft Teams
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 154 | 93.3% | 100.0% |
| Veeva Safety | 7 | 4.2% | 0.0% |
| Microsoft Teams | 3 | 1.8% | 0.0% |
| Notepad | 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 |
|---|---|---|
| Phobos | Veeva Safety | 3 |
| Veeva Safety | Microsoft Teams | 1 |
| Phobos | Microsoft Teams | 1 |
| Phobos | Notepad | 1 |
| Microsoft Teams | Phobos | 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.
Phobos DOM 1,573 events 114 cases 41.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
371×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
264×
pyzeClick
<button>
"Proceed"
212×
pyzeClick
<button>
"Save Comments"
190×
pyzeClick
<button>
"Push to Case Processing"
117×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
75×
pyzeClick
<div>
"Edit info and/or add comments · × · Compound · Select C"
74×
pyzeClick
<button>
"Advance"
64×
Veeva Safety DOM 585 events 96 cases 3.7 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Adverse Events"
38×
pyzeClick
<a>
"Show more"
27×
pyzeClick
<span>
"This Case is Locked"
26×
pyzeClick
<span>
"Products"
19×
pyzeClick
<button>
"Save"
18×
pyzeClick
<li>
"This Case is Locked"
18×
pyzeClick
<li>
"Narrative"
17×
pyzeClick
<li>
"Details"
17×
collaboration.merck.com App 34 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
360 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 24 events 13 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Priorización
186×
file
Daily tracking list (PV-SMART) copy
56×
file
MERCK2519807_ARGPRO_20251126220317
55×
file
MERCK2518114_ARGPRO_20251125190215 (1)
18×
file
Query 21-NOV-2025
9×
file
210-PV007-006 Important Medical Event (IME) List
2×
Microsoft Teams App 23 events 18 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
282 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.
RIM Vault DOM 20 events 4 cases 0.1 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Agreement Documents"
2×
pyzeClick
<div>
"This report requires you to complete the report v"
2×
pyzeClick
<div>
"BDL Agreement to Clinical Study > Clinical Study"
2×
pyzeClick
<span>
"Continue"
2×
pyzeClick
<div>
"Products"
2×
pyzeClick
<a>
"Consortium Agreement NCI MK-3475"
2×
pyzeClick
<span>
"Agreements by Clinical Study"
1×
pyzeClick
<div>
"Clinical Studies"
1×
usc-excel.officeapps.live.com Document 16 events 4 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Acrobat App 16 events 6 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Quality Vault DOM 7 events 5 cases 5.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Microsoft Outlook Message 6 events 5 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.
0.02
Swivel Rate
events in non-primary app / total
0.40
Zero-Edit Visit Rate
inspection-only clicks
1.0
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 0.40
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 0.40
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.00
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.02
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 |
|---|---|---|---|---|---|---|---|
| 2291017 | Analyst 02 | 61.5 | 5.5m | 5 | 0.80 | 0.00 | 1 |
| 2219900 | Analyst 02 | 63.7 | 1.9m | 4 | 0.75 | 0.00 | 1 |
| 2219701 | Analyst 02 | 75.0 | 1.4m | 2 | 0.50 | 0.00 | 1 |
| 2219403 | Analyst 02 | 75.0 | 2955.9m | 2 | 0.50 | 0.00 | 1 |
| 2313396 | Analyst 02 | 75.0 | 0.1m | 2 | 0.50 | 0.00 | 1 |
Users on This Task
7 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) →
09
03
12
01
05
02
13
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 09 | 14 | 3.5 | 97.0 | Revisit Depth | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 09's Dominant Sequence
— 13 of 14 instances
1
Enter MARRS Case ID
This analyst exhibits 2 distinct sequences across their 14 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 24 | 2.2 | 94.4 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 03's Dominant Sequence
— 24 of 24 instances
1
Enter MARRS Case ID
|
||||||||
| ▸ | Analyst 12 | 27 | 3.9 | 93.3 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 12's Dominant Sequence
— 25 of 27 instances
1
Enter MARRS Case ID
This analyst exhibits 3 distinct sequences across their 27 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 29 | 4.1 | 89.7 | Zero-Edit Visits | 0.00 | 0.00 | 4 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 01's Dominant Sequence
— 29 of 29 instances
1
Enter MARRS Case ID
|
||||||||
| ▸ | Analyst 05 | 14 | 3.5 | 88.9 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 05's Dominant Sequence
— 14 of 14 instances
1
Enter MARRS Case ID
|
||||||||
| ▸ | Analyst 02 | 17 | 2.8 | 86.8 | Zero-Edit Visits | 0.21 | 0.00 | 2,973 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 02's Dominant Sequence
— 10 of 17 instances
1
Enter MARRS Case ID
This analyst exhibits 7 distinct sequences across their 17 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 25 | 3.1 | 86.1 | Zero-Edit Visits | 0.00 | 0.00 | 8 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 13's Dominant Sequence
— 25 of 25 instances
1
Enter MARRS Case ID
|
||||||||
94.1
Avg Adherence Score (0-100)
Median 97.5 · Range 71.3–97.5
0.2 min
Avg Task Duration
0.1 min median
1.1
Avg Sub-Steps per Instance
4.7% have multiple sub-steps
—
Reference Path Coverage
190 sub-task instances
Reference Baseline
Categorize Case — Mined SoP
The most common sub-step sequence observed across 190 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Select Report Type
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 192 | 95.0% | 97.2% |
| Veeva Safety | 10 | 5.0% | 2.8% |
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 |
|---|---|---|
| Phobos | Veeva Safety | 7 |
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.
Phobos DOM 2,100 events 143 cases 56.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
403×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
287×
pyzeClick
<button>
"Proceed"
256×
pyzeClick
<button>
"Save Comments"
241×
pyzeClick
<button>
"Push to Case Processing"
118×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
115×
pyzeClick
<button>
"Advance"
103×
pyzeClick
<div>
"Edit info and/or add comments · × · Compound · Select C"
77×
Veeva Safety DOM 1,092 events 128 cases 10.9 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
111×
pyzeClick
<span>
"Adverse Events"
48×
pyzeClick
<a>
"Show more"
47×
pyzeClick
<span>
"Products"
35×
pyzeClick
<li>
"Details"
34×
pyzeClick
<span>
"Narrative"
29×
pyzeClick
<span>
"This Case is Locked"
26×
pyzeClick
<span>
"Case Assessment Results"
26×
RIM Vault DOM 39 events 5 cases 0.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Continue"
4×
pyzeClick
<span>
"Agreement Documents"
3×
pyzeClick
<a>
"MK-2870"
2×
pyzeClick
<div>
"Products"
2×
pyzeClick
<a>
"Consortium Agreement NCI MK-3475"
2×
pyzeClick
<a>
"Kelun"
2×
pyzeClick
<div>
"BDL Agreement to Clinical Study > Clinical Study"
2×
pyzeClick
<div>
"This report requires you to complete the report v"
2×
collaboration.merck.com App 34 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
323 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 34 events 24 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
399 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 26 events 14 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Priorización
182×
file
Daily tracking list (PV-SMART) copy
72×
file
MERCK2519807_ARGPRO_20251126220317
55×
file
MERCK2518114_ARGPRO_20251125190215 (1)
18×
file
Query 21-NOV-2025
9×
file
Book1
3×
file
210-PV007-006 Important Medical Event (IME) List
2×
Acrobat App 20 events 7 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
usc-excel.officeapps.live.com Document 16 events 4 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Word Document 9 events 7 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Quality Vault DOM 8 events 5 cases 5.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.02
Swivel Rate
events in non-primary app / total
0.19
Zero-Edit Visit Rate
inspection-only clicks
1.0
Avg Revisit Depth
max repeats of any activity
0.01
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 0.19
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 0.19
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.00
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.02
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2353955 | Analyst 01 | 71.3 | 0.2m | 3 | 0.33 | 0.37 | 1 |
| 2353954 | Analyst 01 | 72.2 | 0.3m | 3 | 0.33 | 0.34 | 1 |
| 2347125 | Analyst 01 | 74.2 | 6.3m | 3 | 0.33 | 0.28 | 1 |
| 2346744 | Analyst 13 | 75.0 | 0.1m | 2 | 0.50 | 0.00 | 1 |
| 2295005 | Analyst 03 | 75.0 | 0.5m | 2 | 0.50 | 0.00 | 1 |
Users on This Task
7 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
12
13
03
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 | 16 | 4.0 | 96.6 | Revisit Depth | 0.00 | 0.00 | 9 |
|
Reference SoP
1
Select Report Type
Analyst 05's Dominant Sequence
— 15 of 16 instances
1
Select Report Type
This analyst exhibits 2 distinct sequences across their 16 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 26 | 3.7 | 95.8 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Select Report Type
Analyst 12's Dominant Sequence
— 26 of 26 instances
1
Select Report Type
|
||||||||
| ▸ | Analyst 13 | 25 | 3.6 | 95.7 | Revisit Depth | 0.02 | 0.00 | 6 |
|
Reference SoP
1
Select Report Type
Analyst 13's Dominant Sequence
— 23 of 25 instances
1
Select Report Type
This analyst exhibits 3 distinct sequences across their 25 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 46 | 3.5 | 95.4 | Zero-Edit Visits | 0.01 | 0.00 | 7 |
|
Reference SoP
1
Select Report Type
Analyst 03's Dominant Sequence
— 45 of 46 instances
1
Select Report Type
This analyst exhibits 2 distinct sequences across their 46 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 21 | 4.2 | 93.9 | Zero-Edit Visits | 0.00 | 0.00 | 3 |
|
Reference SoP
1
Select Report Type
Analyst 09's Dominant Sequence
— 21 of 21 instances
1
Select Report Type
|
||||||||
| ▸ | Analyst 01 | 37 | 4.6 | 91.2 | Zero-Edit Visits | 0.04 | 0.03 | 9 |
|
Reference SoP
1
Select Report Type
Analyst 01's Dominant Sequence
— 33 of 37 instances
1
Select Report Type
This analyst exhibits 4 distinct sequences across their 37 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 19 | 2.7 | 90.0 | Zero-Edit Visits | 0.03 | 0.00 | 1 |
|
Reference SoP
1
Select Report Type
Analyst 02's Dominant Sequence
— 18 of 19 instances
1
Select Report Type
This analyst exhibits 2 distinct sequences across their 19 instances; the dominant one is shown.
|
||||||||
90.1
Avg Adherence Score (0-100)
Median 97.5 · Range 45.2–97.5
18.9 min
Avg Task Duration
0.1 min median
1.3
Avg Sub-Steps per Instance
15.5% have multiple sub-steps
1.1%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
1.1% of instances follow this exactly
Categorize Case — Mined SoP
The most common sub-step sequence observed across 702 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Enter Case Type
2.3s
Phobos
57%
2
Navigate to Cases
0.0s
Veeva Safety
1%
Variant Explorer
Reference
Categorize Case Reference Path
8 instances · 3 analysts
This is the mined reference path — 8 instances across 3 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Categorize Case.
Genuine Complexity
Complex-Case Execution Path
340 instances · 8 analysts
· Δ adherence +1.2
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
×
Navigate to Cases
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 91.3 vs reference 90.1 (delta +1.2). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
144 instances · 7 analysts
· Δ adherence +5.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
×
Navigate to Cases
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 95.7 vs reference 90.1 (delta +5.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
112 instances · 7 analysts
· Δ adherence +1.9
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter MARRS Case ID
reference at this step: Enter Case Type
×
Navigate to Cases
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 92.0 vs reference 90.1 (delta +1.9). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
20 instances · 5 analysts
· Δ adherence +1.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
2
Enter MARRS Case ID
reference at this step: Navigate to Cases
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 91.1 vs reference 90.1 (delta +1.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
7 instances · 3 analysts
· Δ adherence +1.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter MARRS Case ID
reference at this step: Enter Case Type
2
Select Report Type
reference at this step: Navigate to Cases
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 91.1 vs reference 90.1 (delta +1.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Veeva Safety
4 instances · 3 analysts
· Δ adherence -18.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Start Approval Workflow
reference at this step: Navigate to Cases
Cross-App Swivel — 3 analysts pivot into Veeva Safety 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
3 instances · 3 analysts
· Δ adherence +0.7
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
2
Enter MARRS Case ID
reference at this step: Navigate to Cases
3
Enter Inbox ID
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 3 analysts across 3 cases.
Cross-App Swivel
Cross-App Lookup to Microsoft Teams
3 instances · 3 analysts
· Δ adherence -17.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Microsoft Teams
reference at this step: Navigate to Cases
Cross-App Swivel — 3 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 Veeva Safety
3 instances · 1 analysts
· Δ adherence -22.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Navigate to Home
reference at this step: Navigate to Cases
Cross-App Swivel — 1 analysts pivot into Veeva Safety 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 Teams
2 instances · 1 analysts
· Δ adherence -15.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter MARRS Case ID
reference at this step: Enter Case Type
2
Microsoft Teams
reference at this step: Navigate to Cases
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 Veeva Safety
2 instances · 2 analysts
· Δ adherence -15.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
2
View Case
reference at this step: Navigate to Cases
Cross-App Swivel — 2 analysts pivot into Veeva Safety 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
2 instances · 2 analysts
· Δ adherence +7.4
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
2
Enter Case Type
reference at this step: Navigate to Cases
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 97.5 vs reference 90.1 (delta +7.4). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 748 | 83.2% | 95.8% |
| Veeva Safety | 118 | 13.1% | 2.5% |
| Microsoft Teams | 8 | 0.9% | 0.0% |
| Acrobat | 7 | 0.8% | 0.0% |
| collaboration.merck.com | 5 | 0.6% | 0.0% |
| Microsoft Word | 3 | 0.3% | 0.0% |
| Microsoft Excel | 3 | 0.3% | 0.0% |
| Quality Vault | 2 | 0.2% | 1.7% |
| CelonisTaskMining.Client | 1 | 0.1% | 0.0% |
| usc-excel.officeapps.live.com | 1 | 0.1% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Phobos | Veeva Safety | 52 |
| Phobos | Microsoft Teams | 4 |
| Phobos | collaboration.merck.com | 3 |
| Veeva Safety | Phobos | 3 |
| Microsoft Word | Acrobat | 3 |
| Acrobat | Veeva Safety | 3 |
| Phobos | Acrobat | 2 |
| Veeva Safety | Microsoft Teams | 2 |
| Veeva Safety | Microsoft Word | 2 |
| Veeva Safety | collaboration.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.
Phobos DOM 4,209 events 297 cases 109.1 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
462×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
451×
pyzeClick
<button>
"Save Comments"
328×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
310×
pyzeClick
<button>
"Advance"
216×
pyzeClick
<button>
"Send to process"
208×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
152×
pyzeClick
<a>
"Cases"
145×
Veeva Safety DOM 3,897 events 262 cases 47.7 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
539×
pyzeClick
<a>
"Show more"
180×
pyzeClick
<span>
"Adverse Events"
125×
pyzeClick
<span>
"Narrative"
116×
pyzeClick
<li>
"Narrative"
105×
pyzeClick
<li>
"Details"
90×
pyzeClick
<span>
"Approval Complete"
85×
pyzeClick
<span>
"Products"
82×
Microsoft Word Document 101 events 46 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 538518 (English) (26459147_0_1_344628)
198×
file
V-Narrative 542172 (English) (26517690_0_2_345412)
181×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
file
V-Narrative 538589 (English) (26461195_0_1_346748)
160×
file
V-Narrative 527377 (English) (26286859_0_1_339572)
113×
file
V-Narrative 525025 (English) (26253159_0_3_342205)
109×
file
V-Narrative 537919 (English) (26452534_0_2_341686)
99×
file
V-Narrative 541519 (English) (26506626_0_2_342362)
75×
Acrobat App 99 events 36 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,357 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 90 events 55 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,141 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
collaboration.merck.com App 84 events 40 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
810 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 71 events 34 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Outlook Message 56 events 31 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
RIM Vault DOM 39 events 5 cases 0.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
usc-excel.officeapps.live.com Document 32 events 7 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.05
Swivel Rate
events in non-primary app / total
0.38
Zero-Edit Visit Rate
inspection-only clicks
1.02
Avg Revisit Depth
max repeats of any activity
0.01
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 0.37
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 0.37
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.02
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.05
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2353407 | Analyst 09 | 45.2 | 12.0m | 11 | 0.91 | 0.00 | 5 |
| 2353221 | Analyst 13 | 46.9 | 22.4m | 16 | 0.94 | 0.00 | 4 |
| 2264485 | Analyst 01 | 50.0 | 8.0m | 10 | 0.90 | 0.23 | 1 |
| 2354095 | Analyst 02 | 56.0 | 0.9m | 5 | 0.80 | 0.00 | 2 |
| 2351809 | Analyst 01 | 60.0 | 4.5m | 4 | 0.75 | 0.00 | 1 |
Users on This Task
8 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
10
03
05
12
09
01
13
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 10 | 23 | 2.6 | 93.6 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 10's Dominant Sequence
— 23 of 23 instances
1
Enter Case Type
|
||||||||
| ▸ | Analyst 03 | 151 | 11.6 | 92.2 | Zero-Edit Visits | 0.04 | 0.00 | 60 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 03's Dominant Sequence
— 74 of 151 instances
1
Enter Case Type
This analyst exhibits 15 distinct sequences across their 151 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 56 | 6.2 | 90.7 | Zero-Edit Visits | 0.04 | 0.00 | 10,067 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 05's Dominant Sequence
— 24 of 56 instances
1
Enter Case Type
This analyst exhibits 8 distinct sequences across their 56 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 96 | 10.7 | 90.4 | Zero-Edit Visits | 0.04 | 0.01 | 14 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 12's Dominant Sequence
— 41 of 96 instances
1
Enter Case Type
This analyst exhibits 14 distinct sequences across their 96 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 103 | 6.9 | 90.1 | Zero-Edit Visits | 0.03 | 0.00 | 20 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 09's Dominant Sequence
— 66 of 103 instances
1
Enter Case Type
This analyst exhibits 8 distinct sequences across their 103 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 121 | 9.3 | 89.2 | Zero-Edit Visits | 0.06 | 0.02 | 39 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 01's Dominant Sequence
— 52 of 121 instances
1
Enter Case Type
This analyst exhibits 18 distinct sequences across their 121 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 92 | 8.4 | 88.4 | Zero-Edit Visits | 0.07 | 0.00 | 90 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 13's Dominant Sequence
— 36 of 92 instances
1
Enter Case Type
This analyst exhibits 18 distinct sequences across their 92 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 60 | 7.5 | 86.6 | Zero-Edit Visits | 0.12 | 0.00 | 2,988 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 02's Dominant Sequence
— 24 of 60 instances
1
Enter Case Type
This analyst exhibits 18 distinct sequences across their 60 instances; the dominant one is shown.
|
||||||||
88.0
Avg Adherence Score (0-100)
Median 97.5 · Range 45.2–97.5
25.8 min
Avg Task Duration
0.0 min median
1.3
Avg Sub-Steps per Instance
15.1% have multiple sub-steps
—
Reference Path Coverage
398 sub-task instances
Reference Baseline
Categorize Case — Mined SoP
The most common sub-step sequence observed across 398 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Enter Case Type
2
Navigate to Cases
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 400 | 75.5% | 92.7% |
| Veeva Safety | 101 | 19.1% | 3.5% |
| Acrobat | 7 | 1.3% | 0.0% |
| collaboration.merck.com | 5 | 0.9% | 0.0% |
| Microsoft Teams | 5 | 0.9% | 0.0% |
| Microsoft Word | 3 | 0.6% | 0.0% |
| Microsoft Excel | 3 | 0.6% | 0.0% |
| Quality Vault | 2 | 0.4% | 3.8% |
| CelonisTaskMining.Client | 1 | 0.2% | 0.0% |
| usc-excel.officeapps.live.com | 1 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Phobos | Veeva Safety | 43 |
| Microsoft Word | Acrobat | 3 |
| Phobos | collaboration.merck.com | 3 |
| Acrobat | Veeva Safety | 3 |
| Phobos | Microsoft Teams | 3 |
| Phobos | Acrobat | 2 |
| Veeva Safety | Microsoft Word | 2 |
| Phobos | Microsoft Word | 1 |
| Phobos | Quality Vault | 1 |
| Veeva Safety | Phobos | 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 3,297 events 196 cases 42.0 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
488×
pyzeClick
<a>
"Show more"
159×
pyzeClick
<span>
"Narrative"
105×
pyzeClick
<span>
"Adverse Events"
99×
pyzeClick
<li>
"Narrative"
88×
pyzeClick
<span>
"Approval Complete"
83×
pyzeClick
<span>
"Start Approval Workflow"
71×
pyzeClick
<li>
"Details"
69×
Phobos DOM 3,176 events 229 cases 83.6 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
291×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
181×
pyzeClick
<button>
"Send to process"
171×
pyzeClick
<button>
"Save Comments"
168×
pyzeClick
<button>
"Advance"
159×
pyzeClick
<a>
"Cases"
107×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
105×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
74×
Microsoft Word Document 96 events 41 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 538518 (English) (26459147_0_1_344628)
198×
file
V-Narrative 542172 (English) (26517690_0_2_345412)
181×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
file
V-Narrative 538589 (English) (26461195_0_1_346748)
160×
file
V-Narrative 527377 (English) (26286859_0_1_339572)
113×
file
V-Narrative 525025 (English) (26253159_0_3_342205)
109×
file
V-Narrative 537919 (English) (26452534_0_2_341686)
99×
file
V-Narrative 541519 (English) (26506626_0_2_342362)
75×
Acrobat App 82 events 30 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,254 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 57 events 35 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
791 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Outlook Message 51 events 27 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
subj
Incorrect article uploaded in CLIC 940845
37×
subj
<Due today><Void request><GPVCM POLAND> Case # 2326062
23×
subj
<compensación> Case#2351734
17×
subj
Due today reports on 01-DEC-2025
13×
subj
RE: <compensación> Case#2309077
8×
subj
Untitled
8×
subj
<OT 01-DEC PLEASE REVIEW> Organizacion OT
7×
subj
Due today reports on 27-NOV-2025
4×
collaboration.merck.com App 50 events 29 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Microsoft Excel Document 47 events 23 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
usc-excel.officeapps.live.com Document 16 events 3 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
ONENOTE Document 11 events 6 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.08
Swivel Rate
events in non-primary app / total
0.46
Zero-Edit Visit Rate
inspection-only clicks
1.03
Avg Revisit Depth
max repeats of any activity
0.01
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 0.46
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 0.46
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.03
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.08
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2353407 | Analyst 09 | 45.2 | 12.0m | 11 | 0.91 | 0.00 | 5 |
| 2353221 | Analyst 13 | 46.9 | 22.4m | 16 | 0.94 | 0.00 | 4 |
| 2264485 | Analyst 01 | 50.0 | 8.0m | 10 | 0.90 | 0.23 | 1 |
| 2354095 | Analyst 02 | 56.0 | 0.9m | 5 | 0.80 | 0.00 | 2 |
| 2351809 | Analyst 01 | 60.0 | 4.5m | 4 | 0.75 | 0.00 | 1 |
Users on This Task
8 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
10
03
05
09
01
12
13
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 10 | 23 | 2.6 | 93.6 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 10's Dominant Sequence
— 23 of 23 instances
1
Enter Case Type
|
||||||||
| ▸ | Analyst 03 | 84 | 6.5 | 89.9 | Zero-Edit Visits | 0.06 | 0.01 | 51 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 03's Dominant Sequence
— 74 of 84 instances
1
Enter Case Type
This analyst exhibits 10 distinct sequences across their 84 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 29 | 3.6 | 88.3 | Zero-Edit Visits | 0.09 | 0.01 | 10,057 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 05's Dominant Sequence
— 24 of 29 instances
1
Enter Case Type
This analyst exhibits 4 distinct sequences across their 29 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 71 | 4.7 | 87.8 | Zero-Edit Visits | 0.04 | 0.00 | 16 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 09's Dominant Sequence
— 66 of 71 instances
1
Enter Case Type
This analyst exhibits 4 distinct sequences across their 71 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 62 | 4.8 | 87.7 | Zero-Edit Visits | 0.09 | 0.02 | 26 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 01's Dominant Sequence
— 52 of 62 instances
1
Enter Case Type
This analyst exhibits 11 distinct sequences across their 62 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 51 | 5.7 | 86.6 | Zero-Edit Visits | 0.08 | 0.01 | 10 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 12's Dominant Sequence
— 41 of 51 instances
1
Enter Case Type
This analyst exhibits 8 distinct sequences across their 51 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 47 | 4.7 | 85.9 | Zero-Edit Visits | 0.14 | 0.01 | 76 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 13's Dominant Sequence
— 36 of 47 instances
1
Enter Case Type
This analyst exhibits 12 distinct sequences across their 47 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 31 | 5.2 | 84.4 | Zero-Edit Visits | 0.12 | 0.01 | 14 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 02's Dominant Sequence
— 24 of 31 instances
1
Enter Case Type
This analyst exhibits 8 distinct sequences across their 31 instances; the dominant one is shown.
|
||||||||
91.4
Avg Adherence Score (0-100)
Median 97.5 · Range 75.0–97.5
24.0 min
Avg Task Duration
0.1 min median
1.1
Avg Sub-Steps per Instance
9.7% have multiple sub-steps
—
Reference Path Coverage
124 sub-task instances
Reference Baseline
Categorize Case — Mined SoP
The most common sub-step sequence observed across 124 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Enter MARRS Case ID
2
Microsoft Teams
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 133 | 97.1% | 100.0% |
| Microsoft Teams | 2 | 1.5% | 0.0% |
| Veeva Safety | 2 | 1.5% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Phobos | Veeva Safety | 2 |
| Phobos | Microsoft Teams | 1 |
| Microsoft Teams | Phobos | 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.
Phobos DOM 1,573 events 114 cases 41.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
371×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
264×
pyzeClick
<button>
"Proceed"
212×
pyzeClick
<button>
"Save Comments"
190×
pyzeClick
<button>
"Push to Case Processing"
117×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
75×
pyzeClick
<div>
"Edit info and/or add comments · × · Compound · Select C"
74×
pyzeClick
<button>
"Advance"
64×
Veeva Safety DOM 585 events 96 cases 3.7 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Adverse Events"
38×
pyzeClick
<a>
"Show more"
27×
pyzeClick
<span>
"This Case is Locked"
26×
pyzeClick
<span>
"Products"
19×
pyzeClick
<li>
"This Case is Locked"
18×
pyzeClick
<button>
"Save"
18×
pyzeClick
<li>
"Details"
17×
pyzeClick
<li>
"Narrative"
17×
collaboration.merck.com App 34 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
360 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 24 events 13 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Priorización
186×
file
Daily tracking list (PV-SMART) copy
56×
file
MERCK2519807_ARGPRO_20251126220317
55×
file
MERCK2518114_ARGPRO_20251125190215 (1)
18×
file
Query 21-NOV-2025
9×
file
210-PV007-006 Important Medical Event (IME) List
2×
Microsoft Teams App 23 events 18 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
282 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.
RIM Vault DOM 20 events 4 cases 0.1 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Continue"
2×
pyzeClick
<div>
"Products"
2×
pyzeClick
<a>
"Consortium Agreement NCI MK-3475"
2×
pyzeClick
<div>
"BDL Agreement to Clinical Study > Clinical Study"
2×
pyzeClick
<span>
"Agreement Documents"
2×
pyzeClick
<div>
"This report requires you to complete the report v"
2×
pyzeClick
<span>
"Back to previous page"
1×
pyzeClick
<div>
"Show in Tab"
1×
Acrobat App 16 events 6 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
usc-excel.officeapps.live.com Document 16 events 4 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Quality Vault DOM 7 events 5 cases 5.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Microsoft Outlook Message 6 events 5 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.
0.01
Swivel Rate
events in non-primary app / total
0.38
Zero-Edit Visit Rate
inspection-only clicks
1.0
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 0.38
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 0.38
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.00
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2313396 | Analyst 02 | 75.0 | 0.1m | 2 | 0.50 | 0.00 | 1 |
| 2219701 | Analyst 02 | 75.0 | 1.4m | 2 | 0.50 | 0.00 | 1 |
| 2219403 | Analyst 02 | 75.0 | 2955.9m | 2 | 0.50 | 0.00 | 1 |
| 2346744 | Analyst 13 | 77.5 | 0.2m | 3 | 0.33 | 0.00 | 1 |
| 2279519 | Analyst 13 | 82.5 | 0.0m | 1 | 0.00 | 0.00 | 1 |
Users on This Task
7 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) →
09
12
03
01
05
02
13
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 09 | 11 | 2.8 | 97.5 | Revisit Depth | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 09's Dominant Sequence
— 11 of 11 instances
1
Enter MARRS Case ID
|
||||||||
| ▸ | Analyst 12 | 20 | 3.3 | 94.1 | Zero-Edit Visits | 0.00 | 0.00 | 3 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 12's Dominant Sequence
— 18 of 20 instances
1
Enter MARRS Case ID
This analyst exhibits 3 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 21 | 1.9 | 93.9 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 03's Dominant Sequence
— 21 of 21 instances
1
Enter MARRS Case ID
|
||||||||
| ▸ | Analyst 01 | 23 | 3.3 | 90.3 | Zero-Edit Visits | 0.00 | 0.00 | 3 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 01's Dominant Sequence
— 23 of 23 instances
1
Enter MARRS Case ID
|
||||||||
| ▸ | Analyst 05 | 14 | 3.5 | 90.0 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 05's Dominant Sequence
— 12 of 14 instances
1
Enter MARRS Case ID
This analyst exhibits 2 distinct sequences across their 14 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 10 | 2.0 | 89.2 | Zero-Edit Visits | 0.15 | 0.00 | 2,962 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 02's Dominant Sequence
— 7 of 10 instances
1
Enter MARRS Case ID
This analyst exhibits 4 distinct sequences across their 10 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 25 | 3.1 | 86.8 | Zero-Edit Visits | 0.01 | 0.00 | 8 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 13's Dominant Sequence
— 20 of 25 instances
1
Enter MARRS Case ID
This analyst exhibits 3 distinct sequences across their 25 instances; the dominant one is shown.
|
||||||||
93.9
Avg Adherence Score (0-100)
Median 97.5 · Range 62.2–97.5
0.3 min
Avg Task Duration
0.1 min median
1.3
Avg Sub-Steps per Instance
20.6% have multiple sub-steps
—
Reference Path Coverage
180 sub-task instances
Reference Baseline
Categorize Case — Mined SoP
The most common sub-step sequence observed across 180 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Select Report Type
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 215 | 92.7% | 97.4% |
| Veeva Safety | 15 | 6.5% | 2.6% |
| Microsoft Teams | 1 | 0.4% | 0.0% |
| Notepad | 1 | 0.4% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Phobos | Veeva Safety | 8 |
| Veeva Safety | Microsoft Teams | 1 |
| Veeva Safety | Phobos | 1 |
| Phobos | Notepad | 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.
Phobos DOM 2,100 events 143 cases 56.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
403×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
287×
pyzeClick
<button>
"Proceed"
256×
pyzeClick
<button>
"Save Comments"
241×
pyzeClick
<button>
"Push to Case Processing"
118×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
115×
pyzeClick
<button>
"Advance"
103×
pyzeClick
<div>
"Edit info and/or add comments · × · Compound · Select C"
77×
Veeva Safety DOM 1,092 events 128 cases 10.9 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
111×
pyzeClick
<span>
"Adverse Events"
48×
pyzeClick
<a>
"Show more"
47×
pyzeClick
<span>
"Products"
35×
pyzeClick
<li>
"Details"
34×
pyzeClick
<span>
"Narrative"
29×
pyzeClick
<span>
"This Case is Locked"
26×
pyzeClick
<span>
"Case Assessment Results"
26×
RIM Vault DOM 39 events 5 cases 0.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Continue"
4×
pyzeClick
<span>
"Agreement Documents"
3×
pyzeClick
<a>
"MK-2870"
2×
pyzeClick
<div>
"Products"
2×
pyzeClick
<a>
"Consortium Agreement NCI MK-3475"
2×
pyzeClick
<a>
"Kelun"
2×
pyzeClick
<div>
"BDL Agreement to Clinical Study > Clinical Study"
2×
pyzeClick
<div>
"This report requires you to complete the report v"
2×
collaboration.merck.com App 34 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
323 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 34 events 24 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
399 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 26 events 14 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Priorización
182×
file
Daily tracking list (PV-SMART) copy
72×
file
MERCK2519807_ARGPRO_20251126220317
55×
file
MERCK2518114_ARGPRO_20251125190215 (1)
18×
file
Query 21-NOV-2025
9×
file
Book1
3×
file
210-PV007-006 Important Medical Event (IME) List
2×
Acrobat App 20 events 7 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
usc-excel.officeapps.live.com Document 16 events 4 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Word Document 9 events 7 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Quality Vault DOM 8 events 5 cases 5.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.02
Swivel Rate
events in non-primary app / total
0.18
Zero-Edit Visit Rate
inspection-only clicks
1.01
Avg Revisit Depth
max repeats of any activity
0.01
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 0.18
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 0.18
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.01
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.02
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2353954 | Analyst 01 | 62.2 | 0.5m | 4 | 0.50 | 0.47 | 2 |
| 2291017 | Analyst 02 | 65.0 | 5.5m | 6 | 0.67 | 0.00 | 1 |
| 2219900 | Analyst 02 | 67.5 | 1.9m | 5 | 0.60 | 0.00 | 1 |
| 2353955 | Analyst 01 | 71.3 | 0.2m | 3 | 0.33 | 0.37 | 1 |
| 2347125 | Analyst 01 | 74.2 | 6.3m | 3 | 0.33 | 0.28 | 1 |
Users on This Task
7 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
13
03
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 | 13 | 4.3 | 96.7 | Revisit Depth | 0.00 | 0.00 | 9 |
|
Reference SoP
1
Select Report Type
Analyst 05's Dominant Sequence
— 12 of 13 instances
1
Select Report Type
This analyst exhibits 2 distinct sequences across their 13 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 20 | 2.9 | 96.4 | Revisit Depth | 0.00 | 0.00 | 6 |
|
Reference SoP
1
Select Report Type
Analyst 13's Dominant Sequence
— 18 of 20 instances
1
Select Report Type
This analyst exhibits 3 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 46 | 3.5 | 95.5 | Zero-Edit Visits | 0.01 | 0.00 | 8 |
|
Reference SoP
1
Select Report Type
Analyst 03's Dominant Sequence
— 41 of 46 instances
1
Select Report Type
This analyst exhibits 4 distinct sequences across their 46 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 25 | 3.6 | 95.2 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Select Report Type
Analyst 12's Dominant Sequence
— 18 of 25 instances
1
Select Report Type
This analyst exhibits 3 distinct sequences across their 25 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 21 | 4.2 | 94.0 | Zero-Edit Visits | 0.00 | 0.00 | 3 |
|
Reference SoP
1
Select Report Type
Analyst 09's Dominant Sequence
— 18 of 21 instances
1
Select Report Type
This analyst exhibits 3 distinct sequences across their 21 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 36 | 4.5 | 91.0 | Zero-Edit Visits | 0.05 | 0.03 | 10 |
|
Reference SoP
1
Select Report Type
Analyst 01's Dominant Sequence
— 26 of 36 instances
1
Select Report Type
This analyst exhibits 6 distinct sequences across their 36 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 19 | 2.7 | 88.9 | Zero-Edit Visits | 0.11 | 0.00 | 12 |
|
Reference SoP
1
Select Report Type
Analyst 02's Dominant Sequence
— 11 of 19 instances
1
Select Report Type
This analyst exhibits 7 distinct sequences across their 19 instances; the dominant one is shown.
|
||||||||
90.0
Avg Adherence Score (0-100)
Median 97.5 · Range 45.2–97.5
19.1 min
Avg Task Duration
0.1 min median
1.3
Avg Sub-Steps per Instance
15.8% have multiple sub-steps
1.0%
Reference Path Coverage
Instances following mined SoP exactly
Reference Baseline
1.0% of instances follow this exactly
Categorize Case — Mined SoP
The most common sub-step sequence observed across 702 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Enter Case Type
2.3s
Phobos
56%
2
Navigate to Cases
0.0s
Veeva Safety
1%
Variant Explorer
Reference
Categorize Case Reference Path
8 instances · 3 analysts
This is the mined reference path — 8 instances across 3 analysts follow this exact sequence. Treat as the proposed Standard Operating Procedure for Categorize Case.
Genuine Complexity
Complex-Case Execution Path
338 instances · 8 analysts
· Δ adherence +1.3
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
×
Navigate to Cases
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 91.3 vs reference 90.0 (delta +1.3). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
144 instances · 7 analysts
· Δ adherence +5.6
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
×
Navigate to Cases
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 95.7 vs reference 90.0 (delta +5.6). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
112 instances · 7 analysts
· Δ adherence +2.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter MARRS Case ID
reference at this step: Enter Case Type
×
Navigate to Cases
skipped — this reference step was not executed
Genuine Complexity — this sequence differs from the reference path (1 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 92.0 vs reference 90.0 (delta +2.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
20 instances · 5 analysts
· Δ adherence +1.1
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
2
Enter MARRS Case ID
reference at this step: Navigate to Cases
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 91.1 vs reference 90.0 (delta +1.1). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Genuine Complexity
Complex-Case Execution Path
7 instances · 3 analysts
· Δ adherence +1.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter MARRS Case ID
reference at this step: Enter Case Type
2
Select Report Type
reference at this step: Navigate to Cases
Genuine Complexity — this sequence differs from the reference path (2 sub-steps vs 2 in reference) but shows no strong pain-point signals: adherence score 91.1 vs reference 90.0 (delta +1.0). Likely a legitimate case-complexity path — not an automation candidate, but worth cataloging.
Cross-App Swivel
Cross-App Lookup to Veeva Safety
4 instances · 3 analysts
· Δ adherence -18.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Start Approval Workflow
reference at this step: Navigate to Cases
Cross-App Swivel — 3 analysts pivot into Veeva Safety during this task (0% of task time in non-primary apps, 50% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Linked Remediation Opportunities
Reviewer Sweep
Reviewer Verification Pass
3 instances · 3 analysts
· Δ adherence +0.8
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
2
Enter MARRS Case ID
reference at this step: Navigate to Cases
3
Enter Inbox ID
extra step (reference ends at step 2)
Reviewer Sweep — a viewing-heavy pattern with 3.0 avg sub-steps and zero edits captured. Consistent with a downstream reviewer re-opening the task to verify earlier work rather than progressing new data. 3 analysts across 3 cases.
Cross-App Swivel
Cross-App Lookup to Microsoft Teams
3 instances · 3 analysts
· Δ adherence -17.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Microsoft Teams
reference at this step: Navigate to Cases
Cross-App Swivel — 3 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.
Linked Remediation Opportunities
Cross-App Swivel
Cross-App Lookup to Veeva Safety
3 instances · 2 analysts
· Δ adherence -15.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
View Validation Results
reference at this step: Navigate to Cases
Cross-App Swivel — 2 analysts pivot into Veeva Safety during this task (0% of task time in non-primary apps, 50% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Linked Remediation Opportunities
Cross-App Swivel
Cross-App Lookup to Veeva Safety
3 instances · 1 analysts
· Δ adherence -22.5
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter Case Type
2
Navigate to Home
reference at this step: Navigate to Cases
Cross-App Swivel — 1 analysts pivot into Veeva Safety during this task (0% of task time in non-primary apps, 50% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Linked Remediation Opportunities
Cross-App Swivel
Cross-App Lookup to Microsoft Teams
2 instances · 1 analysts
· Δ adherence -15.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Enter MARRS Case ID
reference at this step: Enter Case Type
2
Microsoft Teams
reference at this step: Navigate to Cases
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.
Linked Remediation Opportunities
Cross-App Swivel
Cross-App Lookup to Veeva Safety
2 instances · 2 analysts
· Δ adherence -15.0
Observed Sequence (vs Reference SoP on left)
differs from reference
matches reference
1
Select Report Type
reference at this step: Enter Case Type
2
View Case
reference at this step: Navigate to Cases
Cross-App Swivel — 2 analysts pivot into Veeva Safety during this task (0% of task time in non-primary apps, 50% of events outside the primary system). Pattern is consistent with data lookup or cross-referencing rather than a system defect — lighter-weight than a Silent Workaround.
Linked Remediation Opportunities
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 748 | 83.0% | 95.8% |
| Veeva Safety | 120 | 13.3% | 2.5% |
| Microsoft Teams | 8 | 0.9% | 0.0% |
| Acrobat | 7 | 0.8% | 0.0% |
| collaboration.merck.com | 5 | 0.6% | 0.0% |
| Microsoft Word | 3 | 0.3% | 0.0% |
| Microsoft Excel | 3 | 0.3% | 0.0% |
| Quality Vault | 2 | 0.2% | 1.7% |
| Microsoft Outlook | 1 | 0.1% | 0.0% |
| Notepad | 1 | 0.1% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Phobos | Veeva Safety | 55 |
| Phobos | Microsoft Teams | 5 |
| Microsoft Word | Acrobat | 3 |
| Phobos | collaboration.merck.com | 3 |
| Acrobat | Veeva Safety | 3 |
| Phobos | Microsoft Excel | 2 |
| Veeva Safety | Phobos | 2 |
| Phobos | Acrobat | 2 |
| Veeva Safety | Microsoft Teams | 2 |
| Veeva Safety | Microsoft Word | 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.
Phobos DOM 4,209 events 297 cases 109.1 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
462×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
451×
pyzeClick
<button>
"Save Comments"
328×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
310×
pyzeClick
<button>
"Advance"
216×
pyzeClick
<button>
"Send to process"
208×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
152×
pyzeClick
<a>
"Cases"
145×
Veeva Safety DOM 3,897 events 262 cases 47.7 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
539×
pyzeClick
<a>
"Show more"
180×
pyzeClick
<span>
"Adverse Events"
125×
pyzeClick
<span>
"Narrative"
116×
pyzeClick
<li>
"Narrative"
105×
pyzeClick
<li>
"Details"
90×
pyzeClick
<span>
"Approval Complete"
85×
pyzeClick
<span>
"Products"
82×
Microsoft Word Document 101 events 46 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 538518 (English) (26459147_0_1_344628)
198×
file
V-Narrative 542172 (English) (26517690_0_2_345412)
181×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
file
V-Narrative 538589 (English) (26461195_0_1_346748)
160×
file
V-Narrative 527377 (English) (26286859_0_1_339572)
113×
file
V-Narrative 525025 (English) (26253159_0_3_342205)
109×
file
V-Narrative 537919 (English) (26452534_0_2_341686)
99×
file
V-Narrative 541519 (English) (26506626_0_2_342362)
75×
Acrobat App 99 events 36 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,357 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 90 events 55 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,141 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
collaboration.merck.com App 84 events 40 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
810 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 71 events 34 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Outlook Message 56 events 31 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
RIM Vault DOM 39 events 5 cases 0.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
usc-excel.officeapps.live.com Document 32 events 7 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.05
Swivel Rate
events in non-primary app / total
0.38
Zero-Edit Visit Rate
inspection-only clicks
1.02
Avg Revisit Depth
max repeats of any activity
0.01
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 0.38
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 0.38
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.02
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.05
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2353407 | Analyst 09 | 45.2 | 12.0m | 11 | 0.91 | 0.00 | 5 |
| 2353221 | Analyst 13 | 46.9 | 22.4m | 16 | 0.94 | 0.00 | 4 |
| 2264485 | Analyst 01 | 50.0 | 8.0m | 10 | 0.90 | 0.23 | 1 |
| 2354095 | Analyst 02 | 56.0 | 0.9m | 5 | 0.80 | 0.00 | 2 |
| 2351809 | Analyst 01 | 60.0 | 4.5m | 4 | 0.75 | 0.00 | 1 |
Users on This Task
8 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
10
03
05
12
09
01
13
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 10 | 23 | 2.6 | 93.6 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 10's Dominant Sequence
— 23 of 23 instances
1
Enter Case Type
|
||||||||
| ▸ | Analyst 03 | 151 | 11.6 | 92.0 | Zero-Edit Visits | 0.04 | 0.00 | 70 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 03's Dominant Sequence
— 73 of 151 instances
1
Enter Case Type
This analyst exhibits 15 distinct sequences across their 151 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 56 | 6.2 | 90.7 | Zero-Edit Visits | 0.04 | 0.00 | 10,067 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 05's Dominant Sequence
— 24 of 56 instances
1
Enter Case Type
This analyst exhibits 8 distinct sequences across their 56 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 96 | 10.7 | 90.4 | Zero-Edit Visits | 0.04 | 0.01 | 14 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 12's Dominant Sequence
— 41 of 96 instances
1
Enter Case Type
This analyst exhibits 14 distinct sequences across their 96 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 103 | 6.9 | 90.1 | Zero-Edit Visits | 0.03 | 0.00 | 20 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 09's Dominant Sequence
— 66 of 103 instances
1
Enter Case Type
This analyst exhibits 8 distinct sequences across their 103 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 121 | 9.3 | 89.2 | Zero-Edit Visits | 0.06 | 0.02 | 39 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 01's Dominant Sequence
— 52 of 121 instances
1
Enter Case Type
This analyst exhibits 18 distinct sequences across their 121 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 92 | 8.4 | 88.3 | Zero-Edit Visits | 0.08 | 0.00 | 183 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 13's Dominant Sequence
— 35 of 92 instances
1
Enter Case Type
This analyst exhibits 19 distinct sequences across their 92 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 60 | 7.5 | 86.6 | Zero-Edit Visits | 0.12 | 0.00 | 2,988 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 02's Dominant Sequence
— 24 of 60 instances
1
Enter Case Type
This analyst exhibits 18 distinct sequences across their 60 instances; the dominant one is shown.
|
||||||||
87.9
Avg Adherence Score (0-100)
Median 92.5 · Range 45.2–97.5
26.0 min
Avg Task Duration
0.0 min median
1.3
Avg Sub-Steps per Instance
15.6% have multiple sub-steps
—
Reference Path Coverage
398 sub-task instances
Reference Baseline
Categorize Case — Mined SoP
The most common sub-step sequence observed across 398 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Enter Case Type
2
Navigate to Cases
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 400 | 75.2% | 92.7% |
| Veeva Safety | 103 | 19.4% | 3.5% |
| Acrobat | 7 | 1.3% | 0.0% |
| collaboration.merck.com | 5 | 0.9% | 0.0% |
| Microsoft Teams | 5 | 0.9% | 0.0% |
| Microsoft Word | 3 | 0.6% | 0.0% |
| Microsoft Excel | 3 | 0.6% | 0.0% |
| Quality Vault | 2 | 0.4% | 3.8% |
| usc-excel.officeapps.live.com | 1 | 0.2% | 0.0% |
| ONENOTE | 1 | 0.2% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Phobos | Veeva Safety | 46 |
| Acrobat | Veeva Safety | 3 |
| Microsoft Word | Acrobat | 3 |
| Phobos | collaboration.merck.com | 3 |
| Phobos | Microsoft Teams | 3 |
| Phobos | Acrobat | 2 |
| Veeva Safety | Microsoft Word | 2 |
| usc-excel.officeapps.live.com | collaboration.merck.com | 1 |
| Phobos | Microsoft Word | 1 |
| Phobos | Microsoft Outlook | 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 3,297 events 196 cases 42.0 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
488×
pyzeClick
<a>
"Show more"
159×
pyzeClick
<span>
"Narrative"
105×
pyzeClick
<span>
"Adverse Events"
99×
pyzeClick
<li>
"Narrative"
88×
pyzeClick
<span>
"Approval Complete"
83×
pyzeClick
<span>
"Start Approval Workflow"
71×
pyzeClick
<li>
"Details"
69×
Phobos DOM 3,176 events 229 cases 83.6 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<button>
"Proceed"
291×
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
181×
pyzeClick
<button>
"Send to process"
171×
pyzeClick
<button>
"Save Comments"
168×
pyzeClick
<button>
"Advance"
159×
pyzeClick
<a>
"Cases"
107×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
105×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
74×
Microsoft Word Document 96 events 41 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
V-Narrative 538518 (English) (26459147_0_1_344628)
198×
file
V-Narrative 542172 (English) (26517690_0_2_345412)
181×
file
V-Narrative 542162 (English) (26518018_0_1_342783)
170×
file
V-Narrative 538589 (English) (26461195_0_1_346748)
160×
file
V-Narrative 527377 (English) (26286859_0_1_339572)
113×
file
V-Narrative 525025 (English) (26253159_0_3_342205)
109×
file
V-Narrative 537919 (English) (26452534_0_2_341686)
99×
file
V-Narrative 541519 (English) (26506626_0_2_342362)
75×
Acrobat App 82 events 30 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
1,254 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Teams App 57 events 35 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
791 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
Microsoft Outlook Message 51 events 27 cases
Message-level capture
Task Mining captures email subjects. Case numbers frequently appear in the subject line.
subj
Incorrect article uploaded in CLIC 940845
37×
subj
<Due today><Void request><GPVCM POLAND> Case # 2326062
23×
subj
<compensación> Case#2351734
17×
subj
Due today reports on 01-DEC-2025
13×
subj
Untitled
8×
subj
RE: <compensación> Case#2309077
8×
subj
<OT 01-DEC PLEASE REVIEW> Organizacion OT
7×
subj
<Due today><Void request><GPVCM POLAND> Case #
4×
collaboration.merck.com App 50 events 29 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
Microsoft Excel Document 47 events 23 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
usc-excel.officeapps.live.com Document 16 events 3 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
ONENOTE Document 11 events 6 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.08
Swivel Rate
events in non-primary app / total
0.46
Zero-Edit Visit Rate
inspection-only clicks
1.03
Avg Revisit Depth
max repeats of any activity
0.01
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 0.46
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 0.46
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.03
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.08
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2353407 | Analyst 09 | 45.2 | 12.0m | 11 | 0.91 | 0.00 | 5 |
| 2353221 | Analyst 13 | 46.9 | 22.4m | 16 | 0.94 | 0.00 | 4 |
| 2264485 | Analyst 01 | 50.0 | 8.0m | 10 | 0.90 | 0.23 | 1 |
| 2354095 | Analyst 02 | 56.0 | 0.9m | 5 | 0.80 | 0.00 | 2 |
| 2351809 | Analyst 01 | 60.0 | 4.5m | 4 | 0.75 | 0.00 | 1 |
Users on This Task
8 analysts executed this task during the pilot. Scatter plots productivity (instances per active day) against adherence (meta score). Top-right = healthy high-volume analysts; bottom-right = firefighters; top-left = careful low-volume; bottom-left = struggling.
Adherence (meta score) →
Productivity (instances / active day) →
10
03
05
09
01
12
13
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 10 | 23 | 2.6 | 93.6 | Zero-Edit Visits | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 10's Dominant Sequence
— 23 of 23 instances
1
Enter Case Type
|
||||||||
| ▸ | Analyst 03 | 84 | 6.5 | 89.7 | Zero-Edit Visits | 0.06 | 0.01 | 60 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 03's Dominant Sequence
— 73 of 84 instances
1
Enter Case Type
This analyst exhibits 10 distinct sequences across their 84 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 05 | 29 | 3.6 | 88.3 | Zero-Edit Visits | 0.09 | 0.01 | 10,057 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 05's Dominant Sequence
— 24 of 29 instances
1
Enter Case Type
This analyst exhibits 4 distinct sequences across their 29 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 71 | 4.7 | 87.8 | Zero-Edit Visits | 0.04 | 0.00 | 16 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 09's Dominant Sequence
— 66 of 71 instances
1
Enter Case Type
This analyst exhibits 4 distinct sequences across their 71 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 62 | 4.8 | 87.7 | Zero-Edit Visits | 0.09 | 0.02 | 26 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 01's Dominant Sequence
— 52 of 62 instances
1
Enter Case Type
This analyst exhibits 11 distinct sequences across their 62 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 51 | 5.7 | 86.6 | Zero-Edit Visits | 0.08 | 0.01 | 10 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 12's Dominant Sequence
— 41 of 51 instances
1
Enter Case Type
This analyst exhibits 8 distinct sequences across their 51 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 47 | 4.7 | 85.6 | Zero-Edit Visits | 0.15 | 0.01 | 169 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 13's Dominant Sequence
— 35 of 47 instances
1
Enter Case Type
This analyst exhibits 13 distinct sequences across their 47 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 31 | 5.2 | 84.4 | Zero-Edit Visits | 0.12 | 0.01 | 14 |
|
Reference SoP
1
Enter Case Type
2
Navigate to Cases
Analyst 02's Dominant Sequence
— 24 of 31 instances
1
Enter Case Type
This analyst exhibits 8 distinct sequences across their 31 instances; the dominant one is shown.
|
||||||||
91.4
Avg Adherence Score (0-100)
Median 97.5 · Range 75.0–97.5
24.0 min
Avg Task Duration
0.1 min median
1.1
Avg Sub-Steps per Instance
9.7% have multiple sub-steps
—
Reference Path Coverage
124 sub-task instances
Reference Baseline
Categorize Case — Mined SoP
The most common sub-step sequence observed across 124 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Enter MARRS Case ID
2
Microsoft Teams
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 133 | 97.1% | 100.0% |
| Microsoft Teams | 2 | 1.5% | 0.0% |
| Veeva Safety | 2 | 1.5% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Phobos | Veeva Safety | 2 |
| Microsoft Teams | Phobos | 1 |
| Phobos | Microsoft Teams | 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.
Phobos DOM 1,573 events 114 cases 41.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
371×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
264×
pyzeClick
<button>
"Proceed"
212×
pyzeClick
<button>
"Save Comments"
190×
pyzeClick
<button>
"Push to Case Processing"
117×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
75×
pyzeClick
<div>
"Edit info and/or add comments · × · Compound · Select C"
74×
pyzeClick
<button>
"Advance"
64×
Veeva Safety DOM 585 events 96 cases 3.7 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Adverse Events"
38×
pyzeClick
<a>
"Show more"
27×
pyzeClick
<span>
"This Case is Locked"
26×
pyzeClick
<span>
"Products"
19×
pyzeClick
<li>
"This Case is Locked"
18×
pyzeClick
<button>
"Save"
18×
pyzeClick
<li>
"Details"
17×
pyzeClick
<li>
"Narrative"
17×
collaboration.merck.com App 34 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
360 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 24 events 13 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Priorización
186×
file
Daily tracking list (PV-SMART) copy
56×
file
MERCK2519807_ARGPRO_20251126220317
55×
file
MERCK2518114_ARGPRO_20251125190215 (1)
18×
file
Query 21-NOV-2025
9×
file
210-PV007-006 Important Medical Event (IME) List
2×
Microsoft Teams App 23 events 18 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
282 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.
RIM Vault DOM 20 events 4 cases 0.1 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Continue"
2×
pyzeClick
<div>
"Products"
2×
pyzeClick
<a>
"Consortium Agreement NCI MK-3475"
2×
pyzeClick
<div>
"BDL Agreement to Clinical Study > Clinical Study"
2×
pyzeClick
<span>
"Agreement Documents"
2×
pyzeClick
<div>
"This report requires you to complete the report v"
2×
pyzeClick
<span>
"Back to previous page"
1×
pyzeClick
<div>
"Show in Tab"
1×
Acrobat App 16 events 6 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
usc-excel.officeapps.live.com Document 16 events 4 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Quality Vault DOM 7 events 5 cases 5.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Microsoft Outlook Message 6 events 5 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.
0.01
Swivel Rate
events in non-primary app / total
0.38
Zero-Edit Visit Rate
inspection-only clicks
1.0
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 0.38
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 0.38
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.00
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2313396 | Analyst 02 | 75.0 | 0.1m | 2 | 0.50 | 0.00 | 1 |
| 2219701 | Analyst 02 | 75.0 | 1.4m | 2 | 0.50 | 0.00 | 1 |
| 2219403 | Analyst 02 | 75.0 | 2955.9m | 2 | 0.50 | 0.00 | 1 |
| 2346744 | Analyst 13 | 77.5 | 0.2m | 3 | 0.33 | 0.00 | 1 |
| 2350890 | Analyst 05 | 82.5 | 0.0m | 1 | 0.00 | 0.00 | 1 |
Users on This Task
7 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) →
09
12
03
01
05
02
13
User-by-User Breakdown
Click any analyst to see their dominant sequence vs the reference SoP.
| Analyst | Instances | Inst / Day | Adherence | Weakest Vector | Swivel | X-App | Min | |
|---|---|---|---|---|---|---|---|---|
| ▸ | Analyst 09 | 11 | 2.8 | 97.5 | Revisit Depth | 0.00 | 0.00 | 1 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 09's Dominant Sequence
— 11 of 11 instances
1
Enter MARRS Case ID
|
||||||||
| ▸ | Analyst 12 | 20 | 3.3 | 94.1 | Zero-Edit Visits | 0.00 | 0.00 | 3 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 12's Dominant Sequence
— 18 of 20 instances
1
Enter MARRS Case ID
This analyst exhibits 3 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 21 | 1.9 | 93.9 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 03's Dominant Sequence
— 21 of 21 instances
1
Enter MARRS Case ID
|
||||||||
| ▸ | Analyst 01 | 23 | 3.3 | 90.3 | Zero-Edit Visits | 0.00 | 0.00 | 3 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 01's Dominant Sequence
— 23 of 23 instances
1
Enter MARRS Case ID
|
||||||||
| ▸ | Analyst 05 | 14 | 3.5 | 90.0 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 05's Dominant Sequence
— 12 of 14 instances
1
Enter MARRS Case ID
This analyst exhibits 2 distinct sequences across their 14 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 10 | 2.0 | 89.2 | Zero-Edit Visits | 0.15 | 0.00 | 2,962 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 02's Dominant Sequence
— 7 of 10 instances
1
Enter MARRS Case ID
This analyst exhibits 4 distinct sequences across their 10 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 25 | 3.1 | 86.8 | Zero-Edit Visits | 0.01 | 0.00 | 8 |
|
Reference SoP
1
Enter MARRS Case ID
2
Microsoft Teams
Analyst 13's Dominant Sequence
— 20 of 25 instances
1
Enter MARRS Case ID
This analyst exhibits 3 distinct sequences across their 25 instances; the dominant one is shown.
|
||||||||
93.9
Avg Adherence Score (0-100)
Median 97.5 · Range 62.2–97.5
0.3 min
Avg Task Duration
0.1 min median
1.3
Avg Sub-Steps per Instance
20.6% have multiple sub-steps
—
Reference Path Coverage
180 sub-task instances
Reference Baseline
Categorize Case — Mined SoP
The most common sub-step sequence observed across 180 task instances. Use the dropdown on the right to compare any deviation to this baseline.
1
Select Report Type
No deviation variants observed for this task in the current cut — every instance follows the reference path.
Application Mix
Which applications analysts touch while executing this task, measured by event count and active dwell time.
| Application | Events | Event Share | Dwell Share |
|---|---|---|---|
| Phobos | 215 | 92.7% | 97.4% |
| Veeva Safety | 15 | 6.5% | 2.6% |
| Microsoft Teams | 1 | 0.4% | 0.0% |
| Notepad | 1 | 0.4% | 0.0% |
Top Cross-App Transitions (within task instance)
Consecutive events that cross application boundaries within a task instance — the signal where swivel-chair patterns live.
| From | To | Transitions |
|---|---|---|
| Phobos | Veeva Safety | 8 |
| Veeva Safety | Microsoft Teams | 1 |
| Veeva Safety | Phobos | 1 |
| Phobos | Notepad | 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.
Phobos DOM 2,100 events 143 cases 56.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
403×
inputld
<select>
"Select Attachment description · Legal · Legal Pre-Sca"
287×
pyzeClick
<button>
"Proceed"
256×
pyzeClick
<button>
"Save Comments"
241×
pyzeClick
<button>
"Push to Case Processing"
118×
pyzeClick
<select>
"Select Compound · - sitagliptin phosphate Tablet (G"
115×
pyzeClick
<button>
"Advance"
103×
pyzeClick
<div>
"Edit info and/or add comments · × · Compound · Select C"
77×
Veeva Safety DOM 1,092 events 128 cases 10.9 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Complete"
111×
pyzeClick
<span>
"Adverse Events"
48×
pyzeClick
<a>
"Show more"
47×
pyzeClick
<span>
"Products"
35×
pyzeClick
<li>
"Details"
34×
pyzeClick
<span>
"Narrative"
29×
pyzeClick
<span>
"This Case is Locked"
26×
pyzeClick
<span>
"Case Assessment Results"
26×
RIM Vault DOM 39 events 5 cases 0.2 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
pyzeClick
<span>
"Continue"
4×
pyzeClick
<span>
"Agreement Documents"
3×
pyzeClick
<div>
"This report requires you to complete the report v"
2×
pyzeClick
<a>
"MK-2870"
2×
pyzeClick
<a>
"Kelun (SKB) MK-2870, MK-2870 (with MK-3475) & MK-"
2×
pyzeClick
<div>
"Products"
2×
pyzeClick
<a>
"Consortium Agreement NCI MK-3475"
2×
pyzeClick
<a>
"Kelun"
2×
Microsoft Teams App 34 events 24 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
399 granular events captured, but no DOM, document, or message-level metadata is available for this application. Instrumenting this app directly with Pyze would unlock element-level detail.
collaboration.merck.com App 34 events 12 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
323 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 26 events 14 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
file
Priorización
182×
file
Daily tracking list (PV-SMART) copy
72×
file
MERCK2519807_ARGPRO_20251126220317
55×
file
MERCK2518114_ARGPRO_20251125190215 (1)
18×
file
Query 21-NOV-2025
9×
file
Book1
3×
file
210-PV007-006 Important Medical Event (IME) List
2×
Acrobat App 20 events 7 cases
App-level capture
Only application name and dwell time available. Direct instrumentation would unlock deeper detail.
usc-excel.officeapps.live.com Document 16 events 4 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Microsoft Word Document 9 events 7 cases
Document-level capture
Task Mining captures the specific file open during work — file names often encode case context.
Quality Vault DOM 8 events 5 cases 5.3 min
DOM-level capture
Pyze JS captures every element interaction — button labels, field IDs, section panels.
Adherence Vectors
Each task instance is scored on four behavioral vectors. Lower values are more adherent to the mined SoP.
0.02
Swivel Rate
events in non-primary app / total
0.18
Zero-Edit Visit Rate
inspection-only clicks
1.01
Avg Revisit Depth
max repeats of any activity
0.01
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 0.18
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 0.18
Investigate why analysts inspect-and-leave without editing — usually missing upstream data or unclear UI affordances.
Revisit Depth
value 1.01
High revisit depth = mid-flow rework. Trace which sub-step is being repeated; usually a data-readiness gap upstream.
Swivel Rate
value 0.02
Reduce cross-app hops by surfacing the secondary-app data inline (API integration or pre-fetched context panel).
Cross-App Time Share
value 0.01
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 |
|---|---|---|---|---|---|---|---|
| 2353954 | Analyst 01 | 62.2 | 0.5m | 4 | 0.50 | 0.47 | 2 |
| 2291017 | Analyst 02 | 65.0 | 5.5m | 6 | 0.67 | 0.00 | 1 |
| 2219900 | Analyst 02 | 67.5 | 1.9m | 5 | 0.60 | 0.00 | 1 |
| 2353955 | Analyst 01 | 71.3 | 0.2m | 3 | 0.33 | 0.37 | 1 |
| 2347125 | Analyst 01 | 74.2 | 6.3m | 3 | 0.33 | 0.28 | 1 |
Users on This Task
7 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
13
03
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 | 13 | 4.3 | 96.7 | Revisit Depth | 0.00 | 0.00 | 9 |
|
Reference SoP
1
Select Report Type
Analyst 05's Dominant Sequence
— 12 of 13 instances
1
Select Report Type
This analyst exhibits 2 distinct sequences across their 13 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 13 | 20 | 2.9 | 96.4 | Revisit Depth | 0.00 | 0.00 | 6 |
|
Reference SoP
1
Select Report Type
Analyst 13's Dominant Sequence
— 18 of 20 instances
1
Select Report Type
This analyst exhibits 3 distinct sequences across their 20 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 03 | 46 | 3.5 | 95.5 | Zero-Edit Visits | 0.01 | 0.00 | 8 |
|
Reference SoP
1
Select Report Type
Analyst 03's Dominant Sequence
— 41 of 46 instances
1
Select Report Type
This analyst exhibits 4 distinct sequences across their 46 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 12 | 25 | 3.6 | 95.2 | Zero-Edit Visits | 0.00 | 0.00 | 2 |
|
Reference SoP
1
Select Report Type
Analyst 12's Dominant Sequence
— 18 of 25 instances
1
Select Report Type
This analyst exhibits 3 distinct sequences across their 25 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 09 | 21 | 4.2 | 94.0 | Zero-Edit Visits | 0.00 | 0.00 | 3 |
|
Reference SoP
1
Select Report Type
Analyst 09's Dominant Sequence
— 18 of 21 instances
1
Select Report Type
This analyst exhibits 3 distinct sequences across their 21 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 01 | 36 | 4.5 | 91.0 | Zero-Edit Visits | 0.05 | 0.03 | 10 |
|
Reference SoP
1
Select Report Type
Analyst 01's Dominant Sequence
— 26 of 36 instances
1
Select Report Type
This analyst exhibits 6 distinct sequences across their 36 instances; the dominant one is shown.
|
||||||||
| ▸ | Analyst 02 | 19 | 2.7 | 88.9 | Zero-Edit Visits | 0.11 | 0.00 | 12 |
|
Reference SoP
1
Select Report Type
Analyst 02's Dominant Sequence
— 11 of 19 instances
1
Select Report Type
This analyst exhibits 7 distinct sequences across their 19 instances; the dominant one is shown.
|
||||||||