How to Run Customer Interviews Before Launching a Local Service
Customer interviews are useful when they reveal how people already solve a problem. A friendly reaction to an idea is weaker evidence than a recent purchase, an inconvenient workaround or a booking. This guide turns five short conversations into a small, testable service offer. The worksheet and examples are Foundlivo planning exercises, not survey results.
Choose one situation to investigate
Start with a customer and a task: for example, apartment residents who arranged a move-out clean during the past six months. Avoid mixing landlords, office managers and homeowners in one tiny sample. Their buying decisions may be different. Write down what you need to learn: who books, what triggers the purchase, how much notice is available and which conditions prevent a job.
Invite people without pitching
Ask permission for a ten-minute conversation. Explain that you are researching the task and are not asking them to buy. Recruit beyond friends where possible, through relevant community groups that permit research requests or people who have actually purchased the service. Do not collect addresses, access codes or payment information. Ask before taking notes or recording; written anonymous notes are usually enough for this exercise.
Use a sequence of concrete questions
- When did you last need this service?
- What happened that made it necessary?
- How did you find and compare providers?
- What information did you need before booking?
- What was included, and what caused extra work?
- What did you do if no provider was available?
- What would you repeat or change next time?
Follow the person's account before introducing your offer. Ask what happened next rather than suggesting that a particular step must have been frustrating. If they cannot remember a price, record it as unknown. Do not turn a guess into a data point.
Keep an evidence log
Use one row per conversation with these columns: anonymous reference, last purchase date, trigger, alternative used, selection criteria, problem, evidence strength and follow-up permission. Separate direct observations from your interpretation. “Two people mentioned uncertainty about included tasks” is an observation. “Everyone wants a premium subscription” is an unsupported leap.
Turn findings into one test
Suppose three interviewees describe confusion over oven cleaning and key handover. A focused test could be a written checklist, an explicit oven option and a confirmed handover window. That does not prove demand. It gives you a specific offer to present to a new group with a real price and actual availability. Count accepted bookings separately from requests for information.
Set a decision rule before the pilot
Choose a rule that fits your capacity, such as continuing only if two suitable customers book and you can deliver within the estimated time. This is your experiment design, not an industry benchmark. Log refusals: wrong location, unsuitable timing, unclear scope and price objections require different changes. If the first test fails, revise one assumption so the next result remains interpretable.
Use outside research appropriately
The U.S. Small Business Administration's planning guidance explains how market research can address demand, location, competition and pricing. Use that broader research alongside your interviews. Five conversations cannot estimate the size of an entire market.
Next, build a clear service scope and a small first-customer test. Written by Celal Bozyel. Published 17 October 2026. General planning information; examples are illustrative.