Most Customer Interviews Are Worthless 🤫

After 8+ years interviewing customers, I keep seeing the same mistakes over and over. Most interviews are broken by design. Not because customers are lying to you, but because we've accidentally created systems that make honesty almost impossible.

What's killing your insights 💀

So here's what I see happening everywhere. People ask super vague questions like "What's your biggest challenge?". It sounds like research but they're actually conversation killers. Too broad, so customers just give you the safest, most generic answers possible.

Or worse, you lead them toward the answer you want. "Don't you think this would save time?" "How frustrating is it when this happens?" You're literally telling them what to say. Even people who genuinely want to help will lean toward agreeing with you.

Then there's this thing I call assumption validation disguised as discovery. You come in with a hypothesis and spend the whole interview trying to prove you're right instead of testing if you might be wrong. If you're not genuinely prepared to hear that you're completely off-base, you won't ask the questions that could reveal that.

People also love asking hypothetical questions. "If we built X, would you use it?" "How much would you pay for Y?" But here's the thing – people are absolutely terrible at predicting their own future behavior. They'll tell you they'd definitely use something, then when you build it, crickets.

And honestly? Even when you do get good insights, half the time they just sit in some beautiful research report that nobody ever looks at again. This happens because either the insights aren't connected to actual business decisions you need to make, or because the person doing the research doesn't have enough influence in the org to actually change anything.

What I do instead ✨

I've had to completely change how I approach these conversations. Now I go in with genuine curiosity instead of trying to validate what I think I already know. I ask customers to explain everything in their own words, even the things I think I understand. Their version is always different from mine.

I also do my homework beforehand to ask questions that actually matter to their world. Instead of "What are your biggest challenges?", I'll say something like "I saw your company just announced that European expansion. How's that affecting your team's priorities right now?" When customers realize you've done the work to understand their context, they're way more likely to go deep with you.

I've started explicitly encouraging contradictory feedback too. I'll say things like "I'm equally interested in what doesn't work" or "Tell me about times when solutions like this have completely failed." This signals that I'm not hunting for validation and actually want the truth, even if it's uncomfortable.

The biggest game-changer has been asking really short, open questions and then just letting them fill in all the gaps. Instead of explaining my whole theory and then asking "Does that resonate?", I'll just say "Walk me through what happened last time you needed to pull together a report for leadership." Way more revealing.

Oh, and I always share something that didn't work for us first. When I admit my failures upfront, customers feel way more comfortable sharing theirs. Creates this psychological safety that leads to actually honest conversations.

One more trick that works really well: instead of asking "Would you buy this?", ask "Why do you think your colleagues might choose this solution?" People project their real feelings onto others without feeling the social pressure to be nice to your face.

How to actually fix your interview process 🔧

Look, if you want to get better at this, here's what I'd do. Start by going back through your last 10 customer conversations. Count how many questions could be answered with just yes or no (spoiler: you want that number to be zero). Notice how often you explained your thinking before asking for their input.

Then work on redesigning how you ask questions. Replace those binary questions with "walk me through..." prompts. Turn leading questions into open exploration. Come up with some vulnerability statements you can use like admitting a feature that flopped or a strategy that didn't work as planned.

Practice the third-person thing in a few conversations. Try focusing on specific recent events instead of asking people to generalize about their behavior. Test out that vulnerability-first approach where you share something that didn't work before asking about their challenges.

After a few weeks of this, start building it into your actual systems. Make sure interview insights are directly connected to product and marketing decisions you're making. Set up regular interview schedules that tie to your roadmap planning. Train your team on these techniques so it's not just you doing it.

The tools that help 🛠️

You don't need fancy expensive software for this. I use Calendly for scheduling. Otter.ai or Fireflies.ai for transcription. Then honestly, ChatGPT or Claude are amazing for helping spot patterns across interviews and synthesizing insights. For real-time collaboration during synthesis, Miro or FigJam work great.

The key is using AI for the mechanical processes like transcription and initial pattern recognition but keeping humans involved for the strategic interpretation and decision-making.

Mistakes I still see experienced people make 🚫

Don't do research just "to understand users better." That's too vague. Start with specific business questions you need answers to. Should we build feature X or Y? Is our pricing too high for mid-market? Which messaging actually resonates with enterprise buyers?

Resist the urge to draw conclusions after just a few conversations. I know it's tempting when you start seeing patterns, but complete your full interview set before you decide what it all means.

Also, don't rely only on interviews. What people say and what they actually do can be completely different. Cross-check your interview insights with analytics, surveys, actual behavior data.

And when you present insights to stakeholders, talk in business language, not research jargon. Connect every finding to something specific they can actually act on.

Conclusion 💡

Customer interviews are either going to be a real competitive advantage for you, or they're just a waste of everyone's time.

The companies that are actually winning in B2B SaaS right now have figured out how to move beyond that superficial validation stuff toward discovery that directly drives their decisions. They've learned to create environments where customers will tell them uncomfortable truths instead of just being polite.

From my experience, the insights that actually change your business are almost never the ones that confirm what you already believed. They're the ones that make you realize you've been asking completely the wrong questions this whole time.

Your customers are totally ready to tell you what you need to know. The question is whether you're ready to actually hear it.

So – what’s next? 👀 How to have customer conversations that don't suck?

Here’s a practical next step: sign up for my free 5-day email series, Customer Conversations That Don’t Suck. One clear lesson and a 20-minute exercise land in your inbox every, so you can try the tactics immediately. In just five days you’ll be running interviews that shape the roadmap instead of sitting in a nicely designed slide deck. You’ll also be able to download the Customer Interview Cheat Sheet as soon as you subscribe on this page.

No cost, no bs, just a straight path to interviews that finally don’t suck 🙂‍↕️