
Customer Interviews: A Practical Guide for New Founders
Ask about real workflows, avoid leading questions, and organize interview notes without overstating demand.
BUSINESS. CAPITAL. BETTER QUESTIONS.
TRACK 03 / LEARN FROM THE WORK
A startup reading track for the messy early questions: Who has the problem? What do they do today? What can you honestly promise next?

A discussion about expanding a sales team may have little to offer when the first customer and deliverable are still unclear. Start with the decision in front of you rather than the milestone that sounds most impressive. This track focuses on learning directly from the work.
Use the interview guide when the task and customer situation need investigation. Move to the first-customers guide when you can describe a bounded offer. Use the experiments guide when you have a specific uncertainty and need to choose a suitable way to observe it.
Ask about the last time someone handled the task. Follow the sequence and the existing workaround. Keep their account separate from their reaction to your proposed product, and do not treat a compliment as an accepted offer.
Preserve differences among participants. A user, a buyer, and an administrator may encounter the same product for different reasons. A small set of conversations can suggest where to investigate next without describing how an entire market behaves.
Define the question, the people involved, the information to collect, and the time or resource limit. Be transparent about prototypes and manual delivery. The point is to learn something useful while meeting any real obligations you create.
After the exercise, record what happened, what you infer, and what remains unknown. You may decide to revise the offer, choose a different customer, or stop a version of the idea. Those are legitimate learning outcomes, not failures to manufacture a positive story.
YOUR NEXT READ

Ask about real workflows, avoid leading questions, and organize interview notes without overstating demand.

Build a focused offer, approach relevant buyers respectfully, and distinguish interest from completed sales.

Translate an idea into a specific question, choose a suitable test, and record what the evidence cannot establish.