One of the most valuable lessons I’ve learned as a Product Manager is that customers are not always good at predicting their own behaviour.

That’s not because they’re misleading us.

It’s because human beings are simply not very good at predicting what they’ll actually do in the future.

A customer might tell you they would use a feature every day.

Then you launch it and usage barely moves.

Another customer might say a feature isn’t important to them.

Six months later, they’re one of your most active users.

This gap between intent and behaviour is one of the most important things Product Managers need to understand.

What Users Say Is Still Valuable

It’s tempting to conclude that customer interviews aren’t reliable because people don’t always behave the way they said they would.

I don’t think that’s the right lesson.

What customers say gives us context that behavioural data can’t provide.

They can explain frustrations.

They can describe goals.

They can tell us what they’re trying to accomplish.

They can reveal problems we wouldn’t discover by looking at analytics alone.

The mistake is treating what they say as a prediction of future behaviour rather than evidence about their needs and motivations.

Behaviour Tells a Different Story

Product analytics show what users actually do.

Which features they use.

Where they drop off.

How frequently they return.

What workflows they complete.

Where they encounter friction.

This is powerful because behaviour happens in the real environment, not in a hypothetical conversation.

A user might say a particular feature is extremely valuable.

But if they’ve never used it after several opportunities, that’s worth investigating.

Perhaps they don’t understand it.

Perhaps it’s difficult to access.

Or perhaps the problem isn’t important enough to change their behaviour.

The data doesn’t tell you why.

But it tells you something is happening.

Intent and Behaviour Should Be Used Together

I’ve learned not to choose between qualitative and quantitative research.

The strongest product insights usually come from combining them.

Imagine your analytics show that users frequently abandon a workflow at the same step.

The behavioural data tells you where the problem occurs.

Customer interviews might reveal why.

Perhaps users don’t understand the terminology.

Maybe they don’t have the information required to continue.

Or they might simply realize that the workflow doesn’t solve the problem they originally came to solve.

Neither data source tells the complete story on its own.

Together, they become much more useful.

The Bigger the Commitment, the More Careful You Should Be

Not all statements about future behaviour deserve the same level of skepticism.

If someone says:

“I’d probably try this feature.”

That’s weak evidence.

If they say:

“I currently spend three hours every week doing this manually, and I’d switch to your product if it solved this problem.”

That’s much more meaningful.

The difference is commitment.

Actual behaviour is strongest when users have something to lose or gain.

That’s why actions like paying, adopting, switching, inviting teammates, or changing an existing workflow are often stronger signals than hypothetical intentions.

Watch What Happens After the Conversation

One habit I’ve developed is following up on customer statements with behavioural evidence.

If a customer says a capability is important, I look at whether similar users actually adopt it.

If someone says onboarding is confusing, I check where users are dropping off.

If customers say they want more advanced functionality, I look at whether existing advanced workflows are already being used.

The goal isn’t to prove customers wrong.

It’s to understand the difference between what they believe they need and what actually changes their behaviour.

That difference can reveal some of the best product opportunities.

Don’t Design for Behaviour You Wish Existed

This is where product teams can get into trouble.

We might believe customers should use a feature because it’s valuable.

So we add banners.

Notifications.

Tooltips.

Pop-ups.

More promotion.

Usage increases temporarily.

But if the underlying problem isn’t important enough, the behaviour rarely lasts.

Instead of asking how to force adoption, ask why users aren’t naturally adopting the capability.

Sometimes the answer is that the product hasn’t created enough value.

Final Thought

I’ve stopped thinking of customer interviews and product analytics as competing sources of truth.

They answer different questions.

What users say helps us understand what they think, want, and experience.

What users do helps us understand what actually happens when they have a choice.

The most effective Product Managers I’ve worked with pay attention to both.

They listen carefully to customers without blindly accepting every prediction.

They study behaviour without treating users like numbers.

Because the real insight often lives in the gap between the two.

And that gap is where some of the most important product decisions begin.