See where customers give up, and why.
Walk your audience through the whole journey, step by step. Every drop-off comes back with the reason behind it, not just the week it happened.
InterviewTake them through the program week by week, and hear at each step what would make them stay or quit.
Benchmarked in public. Trusted by teams at global brands.
Sources · iMario Synthetic Audience Benchmark, public method and raw files · Pew Research Center, survey methodology · Updated 2026-09-05, reviewed by iMario research
Dashboards show the week customers left, never the why. iMario walks the journey with them and brings back the reason, so you fix the cause, not the chart.
Live systems show behaviour, never reasons.
Is what a churn event records. The step is measured to the second; the reason is not a field.
Come back from most of the people who leave. Nobody cancels, they just stop opening it, and the exit survey only reaches the few who bother.
Per hypothesis, because a week-3 test needs enough people to reach week 3 before it can call anything. A wrong guess costs the next one too.
Walk the journey with them, one step at a time.
Lay the flow out, send an audience through it, and hear at every step who is leaving and why. Then rerun the fix on the same people.
Fix the cause, not the chart.
“Week three told me I was behind. I read it as being written off, not encouraged.”
A reason behind every drop
Each week of the curve opens into what people said as they left it, in their own words.
The fix, pre-tested
Rerun the rewritten step on the same cohort before a single real customer sees it.
A journey you can rerun
The whole path stays as an asset, so next quarter’s change is measured against the same people.
Product analytics and a journey read, side by side.
Analytics is the right tool for what happened. This is the tool for why, and for testing the fix before the next cohort reaches the step.
| Decision | Analytics and exit surveys | Journey read on iMario |
|---|---|---|
| What it records | The step and the timestamp. The reason is not a field. | The step, and what the person said at that moment about staying or leaving. |
| Who explains the drop | The few who answer an exit survey. Most people never cancel, they stop opening it. | Everyone who left at that step is still in the audience. Select them and ask. |
| Testing a fix | One quarter per hypothesis, because a real cohort has to reach the step first. | Rewrite the step and rerun the identical cohort the same afternoon. |
| Breaking it down | By the events you instrumented. | By any field the audience carries: age, plan, tenure, region. |
| Cost of a wrong guess | The next intake, and the next quarter. | Credits and an afternoon. |
The week-3 drop nobody could explain.
The message was rewritten and rerun on the same cohort. The curve held before the next intake started.
A twelve-week programme walked end to end by 600 synthetic patients, then the rewritten week rerun on the same cohort — one afternoon.
Where they were
A twelve-week health programme had lost people at week three for a year. Two fixes shipped against two hypotheses; the drop stayed.
Where the notification read as being written off rather than encouraged.
over-60s
Read it that way, against 21% of under-45s — an age split no dashboard showed.
Walking an audience of Synthetic Individuals through a journey step by step, an onboarding, a program or a service flow, and asking at each step whether they continue and why. The result is a retention curve you can open at any step, with the reasons behind each drop.
