A good customer conversation is a small piece of research. Its job is to reveal what happens now, what hurts, and what people already try. Keep the pitch out of the first part of the conversation.
Invite the right people
Choose people in one specific group who have experienced the problem recently. A mix of strangers, acquaintances, and existing connections reduces the chance that politeness dominates the results. Ten is a manageable first batch, not a representative sample.
Ask for fifteen minutes and explain that you are learning about their current process. Do not present the call as a free consultation if you plan to use it as a sales call.
Five useful questions
- Tell me about the last time this happened.
- What did you do to solve it?
- What took the most time or caused the most difficulty?
- What did the workaround cost, including time?
- What would make you change your current approach?
Ask follow-ups such as “What happened next?” and “Can you give me an example?” Avoid leading questions like “Wouldn't an app make that easier?”
Keep a simple evidence log
Record the person's role, recent example, workaround, expense, and an exact phrase that captures the problem. Ask permission before recording audio. Do not collect sensitive details you do not need.
Separate a quoted observation from your assumption. If someone says they use a spreadsheet, do not write that they want new software unless they actually said so.
End without pushing
Thank them and ask whether there is someone else you should talk to. If they ask about your solution, describe the current idea honestly. You can invite them to hear about a pilot later, with their permission.
After the batch, compare patterns. Which problem appeared repeatedly? Who spends money to solve it? Who has authority to buy? If the answers differ by customer type, narrow the group and do another batch.
This week's task: schedule three conversations and use the validation workbook. You can learn more from three concrete stories than thirty compliments.