See churn coming, not in next month's report.
Cohort and churn analysis usually means a spreadsheet someone rebuilds once a month, after the customers have already left. On Finsi, agents watch retention every day, flag who is at risk, and propose retention and win-back campaigns. MAGNET builds the audience, ECHO drafts the email, and your team approves before anything goes out.
Subscription health
Tracks first-order subscription rate (how many new customers subscribe on their first order) and renewal risk points (the days subscribers tend to cancel). When something needs fixing, such as a billing reminder that prompts cancellations, the finding comes with a proposed fix for you to approve.
Cohort analysis
Retention curves, drop-off patterns and revenue per cohort, built every day from verified numbers, not rebuilt by hand each month.
Churn prediction
Models trained on purchase patterns, engagement signals and subscription behavior identify customers at risk before they leave.
Customer health scoring
Every customer gets a health score from recency, frequency, monetary value and engagement, so retention effort goes where it matters most.
Subscription lifecycle
The full subscription journey from first order through renewals, skips, pauses and cancellations, with the moments worth intervening marked.
Churn analysis
Churn reasons broken down by product, price point and experience, so you know which ones drive the most churn.
Retention and win-back proposals
Agents propose retention campaigns, win-backs for lapsed customers, offers and timing based on your data. Each proposal waits for approval, and approved actions are recorded.
Retention cohorts without the spreadsheet.
Retention reports break down customer behavior by cohort, showing when and why customers drop off. Compare cohorts across acquisition channels, products and time periods to see what drives long-term retention.
The full customer lifecycle in one view.
Every customer's journey from first purchase through subscription renewals, with health score, predicted LTV and churn risk in one place, so you know where to focus.
Explore the Platform
More work agents can take off your team
Frequently Asked Questions
1. How does churn prediction work in Finsi?
Finsi uses machine learning models trained on your store's history. They look at purchase recency, order frequency, subscription behavior (skips, pauses, downgrades), engagement and product-level patterns to give each customer a churn probability score, updated as new data comes in.
2. What does cohort analysis show me?
Cohort analysis groups customers by when they were acquired and tracks their behavior over the following months: retention rates, repeat purchase rates, revenue per cohort and subscription survival curves. It shows whether retention is improving and which acquisition channels bring the most loyal customers.
3. What factors determine a customer's health score?
Recency of last purchase, purchase frequency, monetary value (total and average spend), subscription status and tenure, engagement (email opens, site visits) and recent changes such as skipped deliveries or smaller orders. Each factor is weighted by how well it predicts outcomes for your business.
4. How does Finsi track the subscription lifecycle?
Finsi connects to Recharge, Skio, Smartrr and Loop to track activation, renewals, skips, pauses, product swaps, upgrades, downgrades and cancellations. The timeline shows the moments customers are most likely to churn and when intervening makes sense.
5. Do agents contact customers on their own?
No. Agents propose specific actions, such as an offer for an at-risk segment, a re-engagement email, a product swap suggestion or a win-back campaign for recently churned customers, with the target segment, timing and expected impact. A person approves before anything reaches customers, and the action is recorded.
Keep the customers you could save
Tell us how your team tracks retention today. We will show you how agents flag at-risk customers every day and propose what to do, with your team approving each step.