Reading your dashboard
A plain-language guide to each chart on the dashboard: what it shows, why it matters, and how the numbers stay honest.
How to read this page
The dashboard is your one-stop view of churn: how much revenue is at risk, why customers leave, and whether your retention tools are working. Everything here is pulled from real billing data in your Stripe account, nothing is estimated.
The page runs top to bottom:
- Headline numbers. MRR, churn rate, recovered MRR, boosted revenue, and recovery rate.
- Voluntary vs Involuntary Churn. Your churned revenue, split by why customers left.
- Cohort Retention. How long groups of customers keep paying.
- Why Customers Cancel. The reasons people give when they leave.
- Revenue by Cohort. Which cohorts still pay you the most money today.
- Revenue vs Logo Churn. The same churn measured by money versus by customers.
- Save Rate by Plan. How well your cancellation offers work, plan by plan.
Voluntary vs Involuntary Churn
This chart splits your churned revenue into the two very different reasons customers leave:
- Voluntary: the customer decided to cancel. They clicked through a cancel flow and chose to leave anyway.
- Involuntary: a payment failed and the subscription lapsed. The customer didn't choose this; the card did.
Why it matters: the two need different fixes. Involuntary churn is usually a payment problem. Update cards, retry smarter. Voluntary churn is a product or price problem: better offers, better onboarding, better reasons to stay. If you only look at one total churn number, you can't tell which problem to work on.
The customer decides to leave. They go through the cancel flow and choose to cancel anyway.
Fix with: better offers, clearer value, better onboarding.
A payment fails and the subscription lapses. The customer didn't choose this. The card did.
Fix with: card updaters, smarter retries, recovery emails.
Total churned
$641.00
The red line is voluntary churn (customers who chose to leave). The amber line is involuntary churn (failed payments). Here voluntary is higher, meaning better offers and onboarding would save more revenue than smarter payment retries.
Cohort Retention
A cohort is a group of customers who started in the same month. This grid tracks each month's cohort over time and shows what percentage is still subscribed.
Read a row left to right: the first cell is the month they joined (always 100%: everyone who joined is there), and each cell after it is how many of that cohort are still paying one, two, three months later. Green means most of that cohort is still around; red means most have left.
Look for diagonal patterns (a whole month's customers churning fast, which points to a launch problem), horizontal patterns (one cohort doing worse than others, maybe a pricing change that month), and vertical patterns (everyone churning at the same month, a common renewal friction).
Grey cells are months that haven't happened yet. They're blank because we never guess: those numbers simply aren't known yet.
How to spot problems in a cohort grid
- Diagonal: Cohort 1 and Cohort 3 drop sharply and early. Those are usually launch or onboarding problems. These customers never got value in the first place.
- Horizontal: Cohort 3 is weaker than its neighbors across every month. Something about who joined that month (a campaign, a price change) is worth investigating.
- Vertical: If every cohort dropped at the same month, that points to a shared renewal friction rather than a cohort-specific issue.
Each row is a group of customers who started the same month. Read left to right: everyone starts at 100% (M0), then each cell is how many are still subscribed that many months later. Grey cells are months that haven't happened yet.
Why Customers Cancel
Every time a customer goes through a cancel flow, we ask why. This chart groups those answers over the last 30 days and ranks them by how many customers picked each one.
Hover any bar to see the money behind it: how much MRR is at stake for that reason, and the average per attempt. That tells you not just why people leave, but which reasons cost you the most. A rare reason attached to big accounts can hurt more than a common one attached to small plans.
Total cancel attempts
36
Bars are ranked by how many customers gave each reason. Hover a bar to see the MRR at stake. A less common reason attached to big accounts can cost more than a frequent one on small plans.
Revenue by Cohort
The cohort grid counts customers. This chart counts money. Each layer is a start-month cohort, and how tall it stays is how much monthly revenue that cohort still contributes today. Layers are colored by year so you can compare, at a glance, whether this year's customers are sticking better than last year's.
A thick layer that keeps its height is a cohort that still pays you a lot. Customers acquired back then are still generating real revenue. A layer that drops fast is money that walked away early.
Together with the cohort grid, this answers both questions: how many customers stay, and how much money they bring while they do.
The same cohorts as above, but measured in money instead of customers. A layer that stays tall is a cohort still paying you well today; layers are colored by year so you can compare this year's customers against last year's.
Revenue vs Logo Churn
Churn can be measured two ways, and they tell different stories:
- Logo churn: the percentage of customers lost. Every customer counts the same.
- Revenue churn: the percentage of monthly revenue lost. A big account counts more than a small one.
When the two lines diverge, that's the signal. If logo churn is high but revenue churn is low, you're losing many small customers, which is a volume problem. If revenue churn is high but logo churn is low, a few large accounts are leaving, which is a much more expensive problem per customer.
Purple is the share of monthly revenue lost; amber is the share of customers lost. Here logo churn sits above revenue churn. You're losing many small customers rather than a few big ones.
Save Rate by Plan
When a customer tries to cancel, our adaptive offers pick a discount and duration to keep them. A small share of customers always sees your fixed default offer instead, and that is the control group.
This chart shows, plan by plan, what percentage of each group stayed. The brand-colored bars are the adaptive offers; the gray bars are the fixed control. If adaptive bars are taller, the smart offers are genuinely saving more customers than your default would.
Plans with only a handful of sessions can look noisy. Give them time before trusting the gap. A single early win or loss can move a small bar a lot.
Adaptive (bandit offers) vs control (fixed offer) save rate by plan
Brand-colored bars are the adaptive offers; gray bars are the fixed control offer. Taller adaptive bars mean the smart offers are genuinely keeping more customers than your default would.
How the numbers are kept honest
- Everything is real. All numbers come from your actual Stripe account: subscription start dates, prices, cancel events, and failed payments. Nothing is fabricated or estimated.
- Never guess the future. Cohort cells for months that haven't elapsed yet are blank, not extrapolated.
- The 30-day window. Churn-split, reasons, and the churn rates cover the last 30 days so you're looking at current behavior, not ancient history.
- Comparisons use a real control. Save rate is always measured against customers who saw the fixed offer, never against a guess.