Lapsing customer
Trigger
RFM bucket moves from Loyal to At risk
Action
Win-back offer sized by lifetime value; universal holdout excluded
Proof
Incremental revenue against the permanent holdout
Use case · Retention and win-back
Act on the customers about to lapse, before they do.
Retention grids and RFM movement show where a cohort breaks. Predictive segments and computed attributes decide who gets what before the lapse, with a permanent holdout to prove the difference.
Runs on
3 modulesProduct Analytics
The full report suite, from multi-step funnels and cohort grids to RFM, live activity and shared dashboards.
Explore AnalyticsSegmentation & Audiences
Behavioural, demographic and technographic conditions, nested any way, evaluated live, with reachable counts before you launch.
Explore SegmentsAI & Predictive
Copy, segments, reports and journeys from plain language, predictive scores with explanations, and an assistant that never acts without approval.
Explore AI
How it runs
The same play, in the industries where it matters most.
Each example is a trigger on your own events, an action through your own providers, and a number you can hold it to. The industry page shows the rest of that vertical's plays.
FromE-commerce & D2C
FromE-commerce & D2C
Replenishment
Trigger
Computed attribute
days_since_last_ordercrosses the category's median reorder intervalAction
Email with the last order pre-filled, in the customer's language, sent at their best hour
Proof
Repeat-order rate by cohort
Rebooking cadence
Trigger
Computed attribute
median_visit_intervalelapsed sinceappointment_completedAction
Email with the same stylist and service pre-selected, at the customer's best hour
Proof
Rebooking rate by service category
FromMedia, OTT & Subscriptions
Cancellation intent
Trigger
Predictive churn score high and watch time falling
Action
In-app recommendation slot with a curated collection; save offer only if no play in seven days
Proof
Churn lift against holdout
Renewal
Trigger
Subscription ends in fourteen days, engagement score above threshold
Action
Email with progress summary and certificate path; exclude universal holdout
Proof
Renewal lift against holdout
Related industries
Where this use case does the most work.
Each page names the events that matter, the plays that run on them and the reports the team lives on.
FAQ
Questions about retention and win-back.
Anything not answered here is a demo call away.
Request a demoHow does the platform decide who is about to lapse?
Three ways you can combine: RFM movement from one bucket to another, a computed attribute such as days since last order crossing a threshold, and a predictive churn score with the reason behind it.
Do win-back offers go to everyone at risk?
No. The offer is sized by lifetime value or withheld for profiles where it would not pay back, and the universal holdout is excluded from every win-back so incremental revenue is measurable.
Can we see retention before we build anything?
Yes. Retention grids, cohort comparisons and path analysis run on the raw events from the first day of data, so the cliff is visible before a single journey exists.
Get started
See retention and win-back on your own events.
Send your first event in an hour, simulate your first journey against real history, and keep every byte of it in a database you can query.
- No credit card for the Developer plan
- Bring your own providers
- Export everything, any time