Aukiva
All posts

Why Customer Interviews Usually Fail

Tasi Gottschlag · November 12, 2026

Most founders know they should talk to customers, and most of them do. The trouble is that many of these conversations don't teach them much. A founder walks away sure that the need is real and the roadmap is right, builds for a few months, and then watches adoption stay flat.

Usually nobody lied. The interview just wasn't set up to find the truth. In our experience, weak customer conversations are one of the most common reasons companies end up with shaky Product-Market Fit. The problem is rarely how many conversations you have. It's what gets asked in them.

The trap: interviews that only confirm what you already believe

Most founders walk into an interview with an idea already in mind: a product direction, a feature they want to build, or a theory about what customers need. Without meaning to, the conversation turns into an attempt to prove that idea right.

You can hear it in the questions. "Would you use this?" "Does this seem useful?" "Would you pay for something like this?" Most people answer politely and say yes. The founder hears confirmation, and the company learns almost nothing.

The reason is simple: people are bad at predicting what they'll do in the future. Saying "yes, I'd use that" costs nothing. Actually changing how they work, getting budget approved and learning a new tool costs a lot. If you come from engineering, think of it this way: asking a customer "would you use this?" is like testing your code by asking it whether it works. What you want are the logs, a record of what actually happened.

Customers care about results, not features

The second common mistake is spending the conversation on features. Technical founders especially tend to think about what the product does. Customers think about what they're trying to get done: save time, avoid mistakes, close more deals, keep their boss happy.

When the conversation is about features, you learn what customers think of your idea. When it's about results, you learn what they're actually trying to achieve, which tells you what to build and, just as importantly, what to skip. Strong Product-Market Fit comes from delivering the same result for the same kind of customer again and again. It rarely comes from shipping the most requested features.

Ask about the past, not the future

The most useful questions are simple, and they're about things that already happened:

•What problem were you trying to solve?
•What was going on right before you started looking for a solution?
•What else did you try or consider?
•What changed once the problem was solved?
•What would have happened if you'd done nothing?

These questions show you how painful the problem really is, how urgent it was, who made the decision and what result the customer cared about. Because they're based on what people actually did, the answers are far more reliable than guesses about the future.

Here's the difference in practice. A weak question: "Would a dashboard that flags at-risk accounts be useful to you?" Almost everyone says yes. A strong one: "Tell me about the last customer you lost. When did you first notice they were unhappy, and what did you do?" Now you learn whether they even look for warning signs, who owns the problem, and whether it hurts enough to pay to fix.

Habits that make interviews work

Talk mostly to customers who have stayed and gotten real value, plus a few who left. Each group tells you something the other can't. And don't pitch. The moment you start explaining your product, the customer starts reacting to it instead of telling you about their world.

When someone mentions something interesting, ask "Can you tell me about the last time that happened?" A specific story beats a general opinion every time. Write down their exact words, too, because the way customers describe a problem often makes better marketing copy than anything you'll write yourself.

Finally, don't act on any single conversation. One customer asking for a feature is an opinion. The same problem coming up in eight out of twelve conversations is a pattern worth acting on.

Why this matters for Product-Market Fit

When Product-Market Fit is the problem, one of the first things we help companies do is pin down the specific, measurable result customers are buying the product for: the thing they'd genuinely miss if it disappeared. Until that's clear, Product-Market Fit stays shaky. Companies chase too many kinds of customers, try to solve too many problems and build too many features, because every conversation pulls them in a slightly different direction.

That's why good interviews matter so much. They aren't meant to prove you're right. They're meant to show you where you're wrong while it's still cheap to change course. The strongest

SaaS companies go into these conversations curious rather than certain, looking for patterns: which customers succeed, what result those customers value most, and what they have in common. Those patterns are the foundation of Product-Market Fit, and you can start finding them at any stage, whether you have ten customers or a thousand.

Want help working out what your best customers have in common? Book a conversation with Aukiva.

Something limiting your growth?

Book a 30 minute call. No pitch, no pressure.