A tray of fine sand with a straight path of square tiles and a worn diagonal trail of footprints crossing it.

1of 8

Customer interviews that actually discover behaviour

A strong discovery interview starts with a real experience: what happened, how the person tried to solve it and what the workaround still costs them.

Build & Validate
Customer discoveryEvidence-led building

Adapted by the EWOR Team from 2 sources 3 min read

Ask for evidence, not a verdict

A customer interview is not a vote on your idea. Asking whether someone likes your proposed solution risks introducing bias through speculation, social desirability and your own framing. Instead, start with how they experienced the problem. Hold back the solution pitch and reconstruct what they actually did.

Where you have permission, observe their workflow too. What people say they do versus what they actually do often differs, and these observed workarounds may reveal what the interview alone missed.

Start with the right person

The easiest person to interview is not always the right one. Before finding someone to interview, define exactly who you need to speak to, the situation they should have experienced and the problem you want to understand. Then screen for direct experience.

Start by building context without leading them towards an answer:

  • What were you trying to accomplish?
  • When did this last happen?
  • What made it difficult enough to notice?
  • Who else was involved?

If someone has never experienced the situation, they may still have ideas to share, but those ideas are not evidence of how your target users actually behave. Keep that distinction clear rather than treating every conversation as equally valuable.

Start with the problem

When you introduce your product too early, the conversation can turn into a reaction to your explanation rather than an exploration of the problem. Start with the problem, the job the person is trying to get done and the alternatives they use today. Only introduce your concept once you understand that baseline, and treat their reaction as concept feedback, not evidence of actual behaviour.

The goal is not to hide your agenda. It is to understand the customer's world on their terms before your product story shapes the conversation.

Reconstruct one real episode

Move beyond general opinions and get into a specific, recent experience. Follow their first answer with questions that uncover what actually happened:

  1. Trigger: What happened just before the problem appeared?
  2. Sequence: What did you do next, step by step?
  3. Workaround: How do you solve it today?
  4. Consequence: What time, money, risk or frustration did it cost you?
  5. Attempt: What have you already tried to change?
  6. Non-action: If you have not acted, what stopped you?

As you take notes, resist the urge to interpret every answer through your solution. Capture the customer's own words, the sequence of events and the circumstances surrounding their behaviour.

Weigh what you uncover

Not all answers are equally useful. When assessing what you've learned, weigh the evidence from weakest to strongest:

  • A general opinion about the category
  • A prediction about what the person might do
  • A specific event from their past
  • An observable process or workaround they use today
  • A cost, constraint or commitment they've already taken on
  • Repeated behaviour across comparable situations

One interview cannot prove demand, and contradictions matter. A compelling story from the wrong audience, or an expensive workaround unrelated to your intended use case, is not validation. Focus on what the evidence actually tells you, not what you hope it means.

End with what you still do not know

After each conversation, separate what the person told you from what you observed, and keep your interpretation for the synthesis stage. Add a follow-up question that could challenge your current understanding. Across interviews, look for recurring situations and meaningful differences, not just repeated words.

By the end, you should have a clearer picture of who experiences the problem, what happens, how they respond and why it persists. Use those insights to write a focused problem statement, without letting your preferred solution shape the conclusion.

Sources:

No comments yet.

Be the first to comment on this signal.

Get the next signal before everyone else.

Frameworks and contrarian theses for outlier founders, straight to your inbox. No fluff, no recycled advice, weekly.

The EWOR Signals newsletter

Continue learning

Related modules