Two subscription apps can report the same monthly revenue and still be worth very different amounts to a buyer. The difference usually sits in a table most founders rarely open: the cohort view, which follows each month's new subscribers over time. A buyer is pricing the revenue your app will earn after you hand it over, and cohorts are the best evidence you have of what that revenue looks like. Here is how buyers read them, and how to make yours work for you.
1. Total revenue shows where you are. Cohorts show where you are going.
Monthly revenue is a snapshot. It blends subscribers who joined three years ago with subscribers who joined last week, and it can look healthy while the underlying base is getting weaker, or flat while the base is quietly getting stronger. A cohort view separates them. For each starting month, it shows how many subscribers were still paying after one month, three months, a year. That is the closest thing a buyer has to a forecast built from your own users rather than your own hopes, which is why serious buyers ask for it early and lean on it when they set a price.
2. Buyers read cohorts by plan, not as one blended number
A single retention figure hides most of what a buyer wants to know. Monthly and annual subscribers behave differently and renew on different schedules, so a buyer will want them shown separately. They will look at trial conversion, because a trial that converts poorly can make paid acquisition look cheaper than it is. And on annual plans they will look hard at the first renewal, because that is the moment a subscriber decides with their own money whether the app was worth a second year. If your app has annual subscribers who have been through that renewal, that history is one of the most valuable things you can show.
3. Improving cohorts are worth showing, even when revenue is flat
Founders sometimes assume a flat revenue line will be priced as a flat business. Cohorts can change that conversation. If subscribers who joined recently are staying longer than subscribers who joined a year or two ago, that suggests the product is getting better at keeping people, whatever the top line is doing that month. A buyer can underwrite that trend. The same goes for a plan or a market that retains noticeably better than the rest of the app, since that tells a buyer where the next dollar of acquisition spend is likely to work.
4. Separate what the product did from what the pricing did
Cohorts move for reasons that have nothing to do with how much users like the app. A heavy discount promotion brings in subscribers who are more likely to leave at full price. A price increase changes renewal behavior for a cycle or two. A change in where new users come from, such as a burst of paid campaigns, can pull a cohort up or down. A buyer who spots an unexplained dip will usually assume the worst version of the story. A buyer who sees a short written note next to each unusual cohort, saying what happened that month, will usually take the note at face value and move on.
5. Build the view before anyone asks for it
Most subscription apps already have the raw data, whether it lives in RevenueCat, App Store Connect, Google Play Console or Stripe. What is often missing is a consistent version of it. Pick your definitions once: what counts as a new subscriber, how you treat trials, refunds and billing retries, and whether you measure retention by subscribers or by revenue. Then produce the same view for at least the last couple of years, split by plan and platform, and check that it ties back to the revenue a buyer will see in your financials. Doing this before a sale means you find the surprises first, and you get to explain them in your own words.
The takeaway
A buyer is not only paying for the revenue your app earns today. They are paying for how much of it they believe will still be there after you leave. Cohort data is how you show them. Split it by plan, explain the unusual months, and build it before diligence starts, and the strongest parts of your subscriber base will get priced as strengths rather than lost in an average.
