Churn at an all-time high: the retention reckoning

The fastest-growing product I ever worked on was also the one closest to collapse. We just didn’t know it yet. The acquisition numbers were beautiful. New logos every week, pipeline charts climbing in the right direction, and a sales team that had started walking into Monday standups with visible swagger. But underneath those numbers, something was rotting. And nobody wanted to look.

I call this the acquisition illusion. The belief that if the top of the funnel keeps filling, the business must be healthy. It is the most expensive lie in SaaS. And right now, as retention data starts catching up to the growth narrative, more teams are discovering it than at any point in the last decade.

The numbers don’t lie, but they do hide

ProfitWell’s data tells a story that should make every product leader uncomfortable. Churn’s revenue impact is roughly 30% higher than its previous peak during the pandemic boom. But most teams will not feel it yet. They will feel it in six months, when the cohort curves flatten and the board starts asking questions that the acquisition dashboard cannot answer.

Here is what I have learned across twenty years of building products: growth is a narcotic. It feels like progress right up until the withdrawal starts. And the withdrawal, when it comes to churn, does not arrive as a crisis. It arrives as a slow, persistent drag that makes every other number slightly worse, every quarter, until someone finally asks what happened.

But nobody asks what happened when things are going up.

At Freshworks, I watched this pattern play out with the precision of a recurring bad dream. We had strong acquisition. Marketing was performing. Sales was closing. But when someone (finally, reluctantly) pulled the cohort retention data and laid it alongside the acquisition curve, the room went quiet. Accounts that had signed twelve months earlier were disappearing at a rate that made the growth curve meaningless. We were not building a bigger business. We were building a faster treadmill.

The specific moment I remember is a Thursday afternoon meeting where a product lead put two charts on the screen. The first showed monthly new accounts. Beautiful. The second showed net revenue retention by cohort. Devastating. The gap between those two charts was the entire future of the product, and we had been staring at only one of them.

The leaky bucket tax

There is a cost to churn that goes beyond the lost revenue. I call it the leaky bucket tax, and it compounds in ways most product teams never account for.

Every churned customer cost you money to acquire. That is the obvious part. But every churned customer also consumed support resources, onboarding time, engineering bandwidth for feature requests that were specific to their use case, and product roadmap attention that pulled the team away from the customers who were actually going to stay. You paid to bring them in. You paid to serve them. And then you paid for the distortion they introduced into your product decisions on the way out.

But here is the part that really stings. The customers who leave rarely tell you why. They just go quiet. Usage drops. Logins thin out. And then one day someone in finance notices the invoice was not renewed and sends an email that bounces.

I learned this at my first startup. We celebrated every new signup like it was a small victory. We had a Slack channel for it (of course we did). But we had no channel for the users who stopped logging in. No alert, no ritual, no visible signal. The product was haemorrhaging value, and we were too busy watching the front door to notice the back door was wide open.

That startup grew its user count every single month for nine months straight. It also failed. Those two facts are not contradictory. They are causally linked.

Why teams look away

The reason product teams tolerate churn is not ignorance. It is comfort. Acquisition metrics are clean, immediate, and controllable. You can run a campaign, change a price, adjust a landing page, and watch the number move. It feels like agency.

But retention is slow, messy, and humbling. Fixing retention means admitting that the product, the thing you built, is not good enough to keep people. That is a harder conversation than most teams are willing to have. It is certainly a harder presentation to give to a board that has been conditioned to celebrate growth.

Think of a restaurant that fills every table every night. The owner sees the queue out the door and assumes the business is thriving. But if nobody comes back for a second meal, the restaurant is not a restaurant. It is a novelty. And novelty has a half-life.

The most dangerous number in product is the one that only goes up. Because it trains the entire organisation to optimise for more, when the real question is whether enough of the existing base finds enough value to stay.

But teams do not build dashboards for enough. They build dashboards for more.

What retention actually requires

Retention is not a feature. It is not an onboarding flow improvement or a customer success hire (though those things can help). Retention is the product earning the right to be used again tomorrow. And that right is earned in the first fourteen days, reinforced in the first ninety, and lost in the silence between month three and month six when nobody on the product team is watching.

But here is the uncomfortable truth. Most product teams do not own retention. Marketing owns acquisition. Sales owns conversion. Customer success owns expansion. Retention sits in the gap between all three, which means it sits nowhere. And the things that sit nowhere in an organisation are the things that only get fixed after they become emergencies.

I have seen this at three different companies across two continents. The pattern does not change. The org chart determines what gets measured. What gets measured determines what gets fixed. And retention, because it belongs to everyone, belongs to no one.

But the product is the retention mechanism. The experience is the retention mechanism. Every interaction, every loading time, every confusing flow, every moment where a user wonders whether this tool is worth the trouble of opening again tomorrow. That is retention. And that is product work.

The reckoning is not coming. It is here.

The era of growth-at-all-costs produced a generation of products that are excellent at getting people through the door and mediocre at giving them a reason to stay. The data is now catching up. Churn is at levels that make the maths of customer acquisition brutally simple: if you cannot keep them, you cannot afford to get them.

But the teams that figure this out will not be the ones with the best retention dashboards. They will be the ones willing to sit with the uncomfortable question: why are people leaving? And then willing to act on answers that might mean rebuilding the parts of the product that everyone thought were finished.

There is something quietly honest about a product that people come back to without being reminded. No re-engagement email. No discount offer. No push notification dressed up as helpfulness. Just a person opening your product because yesterday it did something worth repeating.

That is the only metric that matters. Everything else is noise wearing a growth costume.