What does a flattening retention curve mean?

A flattening retention curve means your product has found a stable core of users who aren't churning — the curve stops falling and levels off at some percentage above zero. This is one of the strongest signals that your product has real, repeatable value for a segment of your audience. The shape of that curve, and where it flattens, tells you almost everything you need to know about whether you have product-market fit.

The curve itself: what you're actually looking at

A retention curve plots the percentage of users who return to your product over time — day 1, day 7, day 30, day 90, and so on. Almost every product sees an initial drop-off: some users try it once and never return. The critical question is whether the curve keeps falling toward zero or eventually stabilizes at a non-zero baseline.

When the curve flattens — say it stabilizes at 20% retention by day 60 and holds there — it means roughly 20% of your acquired users have made your product a genuine habit. They didn't churn; they found a reason to stay. That residual group is your real market. It is the signal buried inside all the noise of acquisition and early experimentation.

A curve that never flattens and slopes toward zero is an equally important signal: it means no durable habit is forming. Users are sampling your product but not integrating it into their lives. More marketing won't fix that — it will only accelerate your cash burn while the underlying problem remains unaddressed.

Why the level at which it flattens matters as much as whether it flattens

Not all flat curves are equal. A curve that stabilizes at 2% and one that stabilizes at 40% are both technically 'flat,' but they describe entirely different businesses. The level matters because it defines your addressable loyal audience and therefore the ceiling on your LTV.

For consumer apps, flattening anywhere above roughly 25–30% at day 30 is a strong result. For subscription products, the relevant window shifts — you want to see monthly cohorts retaining at a high percentage through the first three to six months before calling it a win. If you're running both monthly and annual subscription tiers, as Rik Haandrikman's work on subscription pricing describes, the flat point on each cohort's curve directly determines whether the annual discount you're offering makes economic sense. A cohort with strong multi-month retention justifies leading with annual; a cohort still finding its flat point suggests you may need monthly pricing to learn what's actually keeping users.

The practical implication: don't compare your flat level against benchmarks from different categories. Compare it against your own previous cohorts. If the curve is flattening higher with each successive cohort, your product improvements are working. If the flat level is stuck or declining, the problem is in the product, not the channel.

What a flat curve unlocks operationally

Once you've confirmed a flat retention curve, three things become actionable that weren't before. First, you can model LTV with real confidence. The flat baseline tells you the minimum revenue per cohort you can expect over a long time horizon, which in turn tells you the maximum you can rationally spend to acquire a user — your CAC ceiling is no longer guesswork.

Second, you can start profiling who your retained users are. The people in that stable residual group almost certainly share traits: a specific use case, a particular onboarding path, a job title or demographic. Understanding what they have in common lets you redirect your acquisition spend toward users who look like them rather than toward users who will inflate early conversion metrics and then churn.

Third, a flat curve gives you the confidence to invest in expansion revenue — upsells, higher tiers, add-ons. Trying to expand revenue from a cohort whose curve hasn't flattened yet is like trying to fill a bucket that still has a hole in it. Once the curve is stable, every incremental improvement to activation or expansion multiplies against a durable base rather than leaking out through churn.

How to move a curve that hasn't flattened yet

If your retention curve is still declining and hasn't found a floor, the instinct is often to run more acquisition experiments. That's almost always the wrong move. The issue is that users aren't finding a reason to return — and solving that is a product problem, not a marketing problem.

The most productive diagnostic is to identify the users who did retain and work backwards. What did they do in their first session that churned users didn't? This is the 'activation moment' question, and it's usually more precise than founders expect. It's often a specific feature used, a specific depth of engagement, or a specific time-to-value threshold crossed.

Once you identify that moment, the job is to pull more users through it faster. That might mean redesigning onboarding, removing steps between signup and first value, or changing how you describe the product in acquisition copy so that the users you attract are already primed for the habit you're trying to build. The goal is a tighter feedback loop between who you acquire and who actually reaches that activation moment — because only users who reach it have any chance of showing up in your flat retention baseline.

“If you're trying to learn... adding a small, secondary monthly option at the bottom won't tell you much at all.”

— Rik Haandrikman, source

The one thing to do

Find the users who are already in your flat retention baseline, identify what they did differently in onboarding, and redesign your activation flow to pull every new user through that same path.

Frequently asked questions

At what retention percentage should a curve flatten to indicate product-market fit?

There is no universal threshold — it depends entirely on your category. Consumer social apps, productivity tools, and subscription services each have different norms. The more reliable signal is trend: if successive cohorts are flattening at higher levels, you're moving in the right direction. A curve that flattens at even 10% is more actionable than a curve still declining at 30%.

Can you have a flat retention curve but still have poor unit economics?

Yes. A flat curve at a very low level, combined with high acquisition costs, can still produce a negative LTV-to-CAC ratio. The curve tells you about product health; your unit economics depend on the level at which it flattens, your pricing, and your cost to acquire those retained users. Both dimensions need to work.

How many cohorts do I need to observe before trusting that my curve has truly flattened?

You generally want to see the pattern hold across at least two to three consecutive cohorts before drawing strong conclusions. A single cohort's flat curve could be an artifact of timing, seasonality, or a one-time acquisition source. Consistency across cohorts is what makes the signal reliable enough to invest against.

Does a flat retention curve mean I should stop worrying about churn?

No. A flat curve means you've found a stable core, but it doesn't mean churn has stopped — it means inflow into that retained segment is roughly matching the slow bleed out. You should still work to push the flat level higher by improving the product experience for existing retained users and by accelerating more new users to the activation moment.

Sources

More playbook answers · Growth Prophet home