Product-Market Fit Validation: Your Sign-Ups Are Growing. Why Aren’t Customers Staying?

Maya was about to spend another month improving an app people kept abandoning.

Her answer to weak retention was already on the roadmap: more recipes, smarter reminders, and a better-looking dashboard.

She had built a meal-planning app for busy professionals. People signed up. Friends loved the idea. Launch comments were encouraging.

But by the following week, most users had disappeared.

Maya thought she had a feature problem.

Before committing to another release, she decided to investigate what happened after someone created an account.

That decision changed the question from “What should we build next?” to “What evidence do we actually have?”

For founders working through product-market fit validation, that question is a useful place to begin.

The gap between liking your idea and needing your product

Maya’s early customer interviews had sounded promising.

People said planning meals was stressful. They wanted to eat better. They liked the idea of having everything organised in one place.

But those conversations left important questions unanswered.

When had they last planned their meals? What had gone wrong? What did they do instead? Had they spent money trying to solve it?

Maya had learned that people recognised the problem. She had not yet established how much it mattered in their lives.

In Stop Building, Start Validating, I explore pain through evidence: workarounds, consequences, urgency, and the time or money customers already spend dealing with a problem.

For your own customer discovery, try asking:

• “Walk me through the last time this happened.”
• “What did you do to deal with it?”
• “What did that cost you?”
• “What have you already tried?” Use the answers to investigate the problem behind the polite agreement.

Her returning users revealed something she had missed

Maya spoke with people who had stopped using the app and people who had returned.

Many of those who left had downloaded it with a general intention to eat healthier. They browsed recipes, saved a few, and carried on ordering food.

The returning users described a more specific struggle.

They were buying groceries without a plan, throwing away unused ingredients, and ordering takeaway when they could not decide what to cook.

They wanted to know what to buy and how to use it before it spoiled.

That gave Maya a narrower target customer and a clearer outcome to investigate.

She revised her value proposition around planning a week of meals using a practical shopping list and overlapping ingredients.

She still needed evidence that the promise worked. A sharper message was the beginning of another test.

How to validate product-market fit through behaviour

Maya’s next experiment was deliberately small.

She invited people with that specific problem to try the weekly planning experience. She watched whether they created a plan, used its shopping list, cooked from it, and returned for another week.

This made her minimum viable product test more useful. Each action answered a question about value.

Creating an account showed that someone was willing to try.

Using the shopping list showed that the product had entered a real shopping trip.

Returning to plan another week suggested that the first experience might have been useful enough to repeat.

One principle from my book matters here: examine behaviour connected to the customer’s outcome.

For your product, identify the action that brings someone closer to the result they came for. Then investigate where people stop before reaching it.

Your product validation process should help you understand whether customers experience the value you promised.

Measure retention around the problem’s rhythm

Maya had been worrying about daily app opens.

But her customers usually planned meals once a week.

She began examining weekly cohorts: groups of users who started during the same period. She compared who completed a plan, who returned the following week, and why some stopped.

That approach reflects another principle in Stop Building, Start Validating: investigate whether the right customers return when the problem returns.

Choose a retention window that matches the job your product serves. Then look beneath the overall percentage.

Which customer segment stays? What do they do differently? What makes the product useful to them? What sends others back to their previous solution?

Customer retention becomes more informative when you understand the behaviour behind it.

Test willingness to pay with a clear offer

Next, Maya offered a paid version of the weekly planning service.

She explained the outcome, what was included, the price, and how to subscribe.

Some people paid. Others declined.

Both responses gave her something to investigate. Did people understand the offer? Had they experienced enough value? Was the problem important enough to justify the price?

For a business that requires customers to pay, willingness to pay belongs in the validation process.

Your pricing experiment needs a specific offer and a real decision. Asking whether a price “sounds reasonable” leaves the financial commitment untested.

Then follow what happens after the first payment. Renewals, cancellations, and the reasons behind them add context to the initial sale.

Bring the evidence together before you scale

Maya’s experiment did not end with a declaration that she had achieved product-market fit.

It gave her better evidence and a clearer next decision.

In my book, the PMF Evidence Scorecard brings together five signals: Signal Question to investigate Pain Does the customer care enough to act? Usage Does the product change meaningful behaviour? Retention Do the right customers return to the value? Payment Do customers commit financially where the business model requires it? Organic pull Do recommendations, invitations, or inbound interest sometimes emerge without individual prompting?

Product-market fit is a body of evidence. Each signal helps you interpret the others.

If customers pay once and disappear, investigate why. If they return but will not pay, examine the offer and business model. If one segment stays while another leaves, study the difference before expanding your audience.

Before you approve another feature, write down the assumption it is supposed to test.

What behaviour would support that assumption? What result would make you reconsider? What is the smallest experiment that could help you decide?

Maya began with a roadmap full of improvements. She moved forward with a clearer understanding of whose problem she was solving and what evidence she still needed.

Your next release deserves that clarity too.