PulseEngagement
Last 30 days
03Product usage

Engagement

What people do after they sign up — which features win, how cohorts retain, and where the system strains.

Feature adoptionWeekly active · 16 weeks
Weekly active users by feature0.60.30.2
DashboardsAlertsReports
Cohort retentionvs. industry benchmark
Weekly cohort retention against the industry benchmark37%
§ 01

Feature rank

Position by weekly active · last 8 weeks
Dashboards#1#1
Alerts#2#3
Reports#3#2
API#4#5
Exports#5#4
Integrations#6#6

Dashboards has held #1 all quarter; Alerts overtook Reports in week 3 and hasn't given the spot back.

§ 02

Under load

Where the system strains when usage climbs
API p95 latencyHourly · milliseconds
API p95 latency with statistical process-control limitsSLO

Two readings breached the ±3σ control band during the 09:00 deploy — inside the 200ms SLO, but worth a look.

Error ratePercent · stepped after deploy
Error rate with a detected regime shift after the deploy+134%

A bad deploy stepped the baseline error rate from ~0.8% to ~1.9%. The regime shift is the story, not any single spike.

§ 03

Who stays

Retention by cohort, and usage per feature
Retention cohortsMonthly · % still active
Monthly signup cohorts by retention ageFebMarAprMayJunJul

The April cohort retains best at equal maturity — it landed just after the guided onboarding shipped.

Feature usageWeekly, per feature
Dashboards
4,020
Alerts
2,030
Reports
1,280
Exports
700
API calls
26,200
Integrations
271