Customer Discovery Interview Questions That Actually Work

The quality of your customer discovery isn't about how many interviews you run — it's about whether you're asking questions that surface real behavior instead of polite agreement. Most founders run ten interviews and learn almost nothing, not because the people were wrong, but because the questions let them be vague.

Here's what to ask instead, and why each question works.

The rule underneath all of this: ask about the past, not the future

People are bad at predicting what they'll do. They're much better at accurately describing what they already did. Every question below is built around that one idea — dig for specific past behavior, not hypothetical future behavior.

Opening: get them talking before you mention your idea

Don't lead with your solution. If you describe your product before you understand their problem, every answer they give you from that point on is contaminated by wanting to be helpful or polite about what you just pitched.

Ask first:

• "Tell me about the last time [problem area] came up for you."
• "Walk me through what you did, step by step."
• "What was frustrating about that?"

Let them talk. Don't fill silences — the most useful information often comes after an initial answer, once they keep thinking out loud.

Digging into the problem itself

Once they've described a real instance of the problem, go deeper:

• "How often does this happen?"
• "What have you tried to fix it?"
• "What didn't work about that?"
• "If you had a magic wand, what would you change about how this works today?"

That last question is useful specifically because it's aspirational but still anchored in a real, already-described problem — you're not asking them to invent a problem, just to imagine relief from one they already have.

Testing how much this actually matters to them

Interest is cheap. Priority and money are not.

• "Where does this rank compared to your other problems right now?"
• "What are you currently paying for, or spending time on, to deal with this?"
• "What would it have been worth to you to have this solved, the last time it happened?"

If someone says the problem is a top priority but they're not spending any money or time on workarounds, that's a signal worth noting — it usually means the pain is real but not yet painful enough to have driven action.

Questions to avoid entirely

• "Would you use a product that does X?" — invites polite agreement, tells you almost nothing
• "Do you think this is a good idea?" — you're asking for their opinion on your idea, not information about their problem
• "How much would you pay for this?" — hypothetical pricing questions produce hypothetical, unreliable answers; a pre-order or waitlist signup is a far better test of willingness to pay
• Leading questions that assume the answer — "Isn't it annoying when X happens?" — people will often just agree with you

Reading the answers

Some patterns worth watching for as you go through interviews:

• Specific, detailed stories — someone who immediately has a vivid recent example is telling you this problem is top-of-mind for them
• Vague or generic answers — "yeah sometimes it's annoying" without a specific story usually means the problem isn't painful enough to have stuck in memory
• Unprompted mentions of workarounds — spreadsheets, sticky notes, other tools cobbled together — this is one of the strongest signals you can get, since it shows they've already invested effort trying to solve it themselves
• Questions they ask you back — "does this exist yet?" or "can I get on a list for this?" is a much stronger signal than any verbal enthusiasm

How many interviews is enough

Ten is usually a reasonable starting point. What matters more than the exact number is whether you're starting to hear the same specific problems and workarounds repeated by people who don't know each other. That repetition is your signal you've found something real rather than a one-off story.

What to do with what you learn

Write down the exact language people use to describe the problem — not your paraphrase of it, their words. That language becomes your landing page copy, your positioning, and often the difference between marketing that resonates and marketing that sounds like every other startup's generic pitch.

Once you've got a handful of interviews with consistent signal, the next step is narrowing that into a scoped first version — see how to validate a startup idea for how to pull it all together.