Why do healthy accounts still churn?

Because your health score is measuring activity, not outcomes, and the renewal was mostly decided during onboarding, long before anyone scheduled the renewal call. A green account tells you the customer is logging in, filing tickets, and showing up to the QBR. It does not tell you whether they got the result they bought. Those are two different questions, and only one of them predicts a renewal. So when a healthy-looking account churns anyway, it is almost never a shock that happened at renewal. It is a decision the customer made in the first ninety days that stayed invisible on your dashboard until the invoice came due.

There is a name for the version of this that runs inside a customer success org. Ghost motion. Retention activity that looks like health and produces nothing you can reproduce. QBRs on the calendar, health scores green, CSMs slammed, and no one can tell you which of it saved which renewal. Green account, churned anyway.

What is a health score actually measuring?

Usually logins, ticket volume, feature adoption, and whether the QBR happened on time. All of that is motion. None of it is outcome. A health score built on usage tells you the customer is busy inside your product. It cannot tell you the customer is winning with it.

Those two come apart far more often than anyone wants to admit. A team can be in the app every single day and still be nowhere near the result they signed up to get. That gap is invisible on a usage chart and fatal at renewal. If your health score is really a usage score, you are grading motion and calling it value. It is the same trap as bolting AI onto a go-to-market motion you can't explain and getting more confusion, faster. Same disease, customer-success badge on it. The full pattern is at ghost motion.

Why was the renewal already decided at onboarding?

Because value realization is the whole game, and it gets won or lost early. A customer who reaches first value fast renews and expands almost on autopilot. A customer who never quite got there was gone the day they quietly stopped believing the tool would pay off. They don't announce it. They don't churn on day 30. They ride out the contract, skip a QBR or two, stay "green" because the license is still provisioned, and then decline to renew ten months later, in a quarter where your onboarding metrics looked perfectly fine.

That is why renewal-quarter heroics so rarely work. By the time a renewal is sixty days out, you are not saving the account. You are discovering a call the customer made three quarters ago and finally getting the bill for it. The window to change the outcome closed while everyone was watching the wrong ninety days.

What should you measure instead of logins?

Two numbers, and together they are the scoreboard that matters. Gross revenue retention and net revenue retention. GRR is the revenue you keep from your existing book before you sell anything new, so it is the cleanest read on whether customers are actually getting value and staying. NRR adds expansion on top. Above 100 percent NRR means your existing customers grow on their own. Below it means you are refilling a leaky bucket before you count a single new logo.

Here is the honest benchmark for a company at your stage, not an enterprise number lifted from someone's deck. The SaaS Capital 2026 benchmark, drawn from more than a thousand private B2B SaaS companies, puts bootstrapped businesses in the $3 to $20M ARR band at a median NRR around 103 percent and a median GRR around 91 percent, with the top decile pushing NRR close to 118 percent. KeyBanc's 16th annual Private SaaS Survey, from late 2025, has net retention holding above 100 percent and gross retention approaching 90 percent across the wider market. Read those two figures together and the message is blunt. The median company loses roughly nine cents of every dollar before it expands anything, and most of that nine cents is the healthy-looking accounts that were never healthy where it counted.

How do you catch a green-but-churning account in time?

You move the whole retention motion upstream, to onboarding, where the outcome actually gets set. Three moves.

Define first value concretely, per segment, and instrument it. Not "logged in." The specific thing a customer does that predicts they will stay, and how fast they reach it. Measure time-to-first-value the way sales measures cycle time. Second, replace usage-based health with outcome-based health. Ask whether the customer hit the result they bought, not whether they opened the app this week. Third, catch risk on a signal, not a calendar. The teams running NRR well don't find out an account is soft at the quarterly review. They flag it sixty days before it shows up, while there is still time to act, instead of documenting it thirty days after the damage is done. Get that right and you have turned retention into a revenue engine instead of a defense.

I ran a customer success org to 100 percent renewal across 168 accounts at an 89 NPS. That number did not come from renewal-quarter saves. It came from a documented onboarding handoff, health signals tied to outcomes instead of logins, and at-risk accounts flagged early enough to actually move them. You can't reproduce what you can't explain, and a green dashboard explains nothing.

The GTM Signal Check includes a CS health read that shows where your retention is really set, in the first ninety days or not at all. Sixty minutes, no pitch, no deck.

Run the GTM Signal Check before your next renewal forecast.

By Eric Glass, Founder, HG Digital. HG Digital is a full-stack GTM firm for growth-stage B2B: we do the GTM work, you run the business.