Funnelsight is a fictional company. This site was built as a training project and does not offer a real product.

Expansion or churn: reading the signals in an account

Updated

An account drops from 25 seats to 18. The team may be cleaning up before it leaves, or settling into steady use after a busy rollout, and the usage data looks the same in both cases.

The same curve, different stories

Each line below follows one feature in one account over the past year. On usage alone, the first two curves look alike, with use that stops. Read with the account's history and its contract, one is expected and the other is a warning.

One feature in each of three customer accounts, monthly use over the last 12 months. Illustrative data
Feature, one account eachMonthly useWhat it signals
Import from the previous toolA peak in the first month, then almost nothingNothing to worry about. A migration happens once, so the drop is expected.
Monthly client report, annual contractUsed every month for ten months, then not at all for the last twoChurn risk. The report stopped in the two months before renewal.
Automated alertsSteady growth every month, up to the plan's limitUpsell. Use has grown every month and reached the plan's limit.

The first account came from a competitor with history to import, the second is about to renew, and the third pays for a plan with a limit. None of that sits in the product: the renewal date is in the CRM and the plan's limits in billing.

Signals, and what can fool them

Each signal below has a usual reading and a common false alarm. Some point to growth, others to a customer drifting away.

One pattern deserves its own watch: customers who buy at the end of a trial without having really used the product, then cancel soon after their first payment. Paying didn't mean they had adopted it.

What tells a pause from a departure

No single signal settles it, which is why many teams combine several into a customer health score. Compare each account with its own history rather than with an average, read several signals together, and check them against the contract and the renewal date. The last piece is what the customer success manager knows. An account that slows down while it waits for a new integrator looks exactly like one that's about to leave, until someone writes it down. Below a certain contract size, many accounts have no customer success manager at all, and the signals are the only view anyone has of them. In Funnelsight, that note sits on the period it explains, and the rules your team sets turn usage, contract and notes into churn-risk and expansion flags in the CRM.

Calibrate on the accounts you already know

Before setting a threshold, look at what the accounts that upgraded did in the two or three months before, and the same for the accounts that left. Keep the signals that separated them, and drop the ones that looked the same in both groups. Check again after a change in pricing, packaging or the product, since a signal that held last year may not hold now.

See usage next to each account's plan, renewal date and notes, in one view.

Start free trial
What's the difference between an expansion signal and a PQL?

A product qualified lead points to a new deal: someone using the product enough to justify a sales conversation. An expansion signal points to an existing customer outgrowing its plan. The tracking logic is similar, but the people involved and the follow-up are not.