Why customers left in Q3
64 accounts cancelled between July and September. Sam called or surveyed every one of them. Claude drew the charts.
Monthly churn, January to September
Share of paying accounts that cancelled each month.
The reason each account gave
One reason per account, in their words, sorted into six piles.
Are newer customers sticking around?
Share of each month's new accounts still paying, by months since they signed up. Darker is better.
| Signed up | Month 1 | Month 2 | Month 3 | Month 4 | Month 5 |
|---|---|---|---|---|---|
| April58 | 88% | 81% | 77% | 74% | 72% |
| May61 | 87% | 80% | 76% | 73% | |
| June55 | 86% | 80% | 75% | ||
| July63 | 89% | 84% | |||
| August66 | 93% |
What we do about it
Setup is the leak
21 of 64 never finished setup. The August checklist already lifted first-month retention from 86% to 93%. Bet 1 in the Q4 roadmap finishes the job.
We lose to the accountant
14 left because their accountant wanted the numbers in another shape. Not one of them named a competitor. That is bet 3.
Price is a small-account problem
All 11 price complaints came from accounts sending fewer than 5 invoices a month. A lighter plan might keep them. Not decided.
This is the final version. The earlier finals are in the trash, where Sam can visit them.