Independent prototype using public U.S. Government help content and synthetic telemetry. Not affiliated with or endorsed by the Department of Defense, the Defense Health Agency or PEO DHMS. No PII or PHI.

Screen 2 of 5 · Instrument the go-live

Go-live radar

Instead of counting tickets after the fact, the radar watches every new user's first weeks: did they sign in, did they finish a real task, did self-service resolve their question, are they stuck? Stuck users trigger proactive nudges.

Synthetic telemetry · fictional rollout

Synthetic data

A fictional product release rolled out to 8 fictional sites over 42 days. Every number on this page comes from a seeded simulation (seed 20261104), so it is the same on every visit. No real users, sites or systems.
Adoption
87%
2,760 of 3,174 invited signed in
Time to first successful task
10.9 h
median, after first sign-in
Self-service containment
53%
questions resolved without a person
Human-touch tickets
20.2
per 100 new users
Sentiment proxy
3.43 / 5
in-app pulse, n = 1,508
Stuck users
1,363
43% of invited; 758 nudges sent

Adoption curve

Cumulative users by day since the first site went live.

01,2502,5003,7505,00017131925313742Day
By day 42, 3,174 users invited, 2,760 signed in and 2,411 completed a first real task.
Show data table
Adoption curve
DayInvitedSigned inFirst successful task
1492416
21045952
3217118102
4308193167
5463291250
6586403348
7693501446
8840606527
9957701612
101,121824713
111,257950822
121,4461,095940
131,5681,2291,057
141,7041,3481,161
151,8631,4661,253
162,0031,5801,363
172,1701,7001,466
182,3341,8311,581
192,4531,9621,690
202,5672,0691,801
212,6722,1721,892
222,7512,2561,958
232,8152,3492,036
242,8692,4242,102
252,9202,4692,138
262,9612,5072,178
272,9932,5462,214
283,0242,5862,252
293,0512,6162,281
303,0692,6412,298
313,0862,6642,318
323,1002,6842,337
333,1172,6922,343
343,1272,7092,361
353,1362,7152,368
363,1422,7222,373
373,1492,7292,381
383,1552,7402,392
393,1642,7462,396
403,1672,7502,401
413,1702,7572,408
423,1742,7602,411

Time to first successful task

Hours from first sign-in to the first completed task. 'Not yet' users are candidates for a walkthrough nudge.

289< 4 h10124-12 h66212-24 h4191-3 days30> 3 days348Not yet
Median 10.9 h. 348 signed-in users have not completed a task yet.
Show data table
Time to first successful task
BucketUsers
< 4 h289
4-12 h1012
12-24 h662
1-3 days419
> 3 days30
Not yet348

Proactive nudges: rules engine

Edit a threshold and every count on this page updates. Treatment users get the nudge; control users are flagged but held, so the experiment stays clean.

OnRuleThresholdNudgeMatched (sent / held)
Invited but never signed inDays since invitedaysReminder with a one-page sign-in guide and a link to the assistantIn-app154 / 251
Signed in, no successful task yetHours since first sign-in without a completed taskhoursGuided walkthrough of the top task for their segmentIn-app119 / 228
Repeated failed attemptsFailed task attemptsattemptsOpen the assistant with the matching article pre-loadedAssistant82 / 184
Active, then went quietDays since last activitydaysShort check-in: what got in the way? (one tap)In-app330 / 333
Low sentiment pulseSentiment pulse at or belowof 5Proactive outreach from a person with the context packet attachedPerson73 / 124
758 nudges sent to treatment users; 1,120 held for control users.

Stuck users (first 10)

Synthetic IDs only. Each row says why the user is stuck and what happens next.

UserWhyNext step
UA0000patient, controlActive, then went quietHeld (control cohort)
UA0001clinician, controlRepeated failed attemptsHeld (control cohort)
UA0002patient, treatmentActive, then went quietShort check-in: what got in the way? (one tap)
UA0004staff, treatmentActive, then went quietShort check-in: what got in the way? (one tap)
UA0006staff, treatmentActive, then went quietShort check-in: what got in the way? (one tap)
UA0008patient, controlInvited but never signed inHeld (control cohort)
UA0018patient, controlInvited but never signed inHeld (control cohort)
UA0020patient, treatmentActive, then went quietShort check-in: what got in the way? (one tap)
UA0024clinician, treatmentActive, then went quietShort check-in: what got in the way? (one tap)
UA0025clinician, treatmentSigned in, no successful task yet; Active, then went quietGuided walkthrough of the top task for their segment

By site

Go-live day, adoption and care load per fictional site (both cohorts).

SiteLiveAdoptionFirst taskContainedTickets /100Stuck
Site A (Juniper Ridge)Day 097%10.9 h53%18.6186
Site B (Cedar Bay)Day 296%10.1 h50%20.0160
Site C (Pinecrest)Day 485%10.5 h56%20.1240
Site D (Saltmarsh)Day 769%9.7 h43%26.8199
Site E (Red Mesa)Day 982%12.6 h53%20.5184
Site F (Northfield)Day 1195%11.9 h55%20.4105
Site G (Lakeside)Day 1481%11.0 h56%17.6177
Site H (Granite Falls)Day 1689%10.4 h52%19.0112