OpenBuilder

Customer Interviews in African Markets: A Practical Guide

Good interviews uncover behaviour, constraints, trust and real alternatives. This guide helps founders learn without leading the customer.

Put the guide into practice

By Innocent Ociti

Published · Updated

Customer interviews are not sales calls and they are not requests for encouragement. They are structured conversations that help a founder understand what people do, why they do it, and what makes change difficult.

In many African markets, the interview must also account for trust, informality, shared devices, cash flows, language, infrastructure, and the influence of family or community. Those factors are not side notes. They often shape the product.

What is the goal of a customer interview?

The goal is to replace assumptions with specific evidence about past and present behaviour.

You want to learn:

  • when the problem last occurred;
  • who was involved in solving it;
  • what the person tried;
  • what the workaround cost in money, time, risk, or dignity;
  • what prevented a better solution;
  • who influences or approves a purchase;
  • what would make the person trust a new option.

Recruit for experience, not enthusiasm

Interview people who recently experienced the situation you are studying. A person can like your idea and still know little about the problem.

Write a short screener with factual questions. For example:

  • When did you last send stock to another town?
  • How do you currently record customer payments?
  • Have you paid for help with this in the past three months?

Avoid recruiting only friends. Ask each relevant participant to introduce you to one other person with a different experience. Community organisations, trade groups, cooperatives, professional associations, schools, clinics, and local businesses can also help you reach the right participants.

Ask about the last time

Hypothetical questions produce optimistic answers. Specific memories produce useful detail.

Instead of “Would you use an app that tracks inventory?” ask:

  1. Tell me about the last time stock went missing.
  2. How did you notice?
  3. What did you do next?
  4. Who else became involved?
  5. What did the incident cost?
  6. What have you tried to prevent it?
  7. Why did that approach continue or stop?

Do not introduce your solution until you understand the current reality.

Listen for systems around the user

The customer may not be the payer. The payer may not be the daily user. A manager, spouse, procurement officer, field agent, regulator, or community leader may influence the decision.

Map five roles:

  • person experiencing the problem;
  • person using the solution;
  • person paying;
  • person approving;
  • person who can block adoption.

If one person holds several roles, record that too.

Design for trust

People may tell you what they think you want to hear, especially when they believe the interviewer controls access to funding, jobs, or services. Reduce that pressure.

Explain that there are no right answers, you are testing your own assumptions, and honest criticism is useful. Do not promise support you cannot provide. Ask permission before recording. Offer another method—notes, voice messages, or a shorter follow-up—when connectivity or time is limited.

Language matters. If the conversation works better in a local language, use it and translate the meaning carefully rather than forcing technical English.

Capture evidence consistently

Immediately after each interview, record:

  • participant segment and context;
  • strongest direct observation;
  • current workaround;
  • frequency and severity;
  • money or time spent;
  • trust or adoption barrier;
  • exact contradiction to your assumptions;
  • next question to investigate.

Do not reduce an interview to “positive” or “negative.” The valuable result is a clearer model of reality.

Synthesize patterns without pretending certainty

After five interviews, compare answers. After ten, look for repeated situations, not just repeated phrases.

Group observations into:

  • common triggers;
  • common workarounds;
  • costly consequences;
  • differences between segments;
  • decision and trust factors;
  • unresolved contradictions.

A pattern is stronger when it appears in behaviour and cost, not only opinion.

End with a next experiment

Every interview round should change what you do. You may narrow the customer segment, rewrite the problem, test a manual service, change the channel, or stop pursuing an assumption.

The purpose of research is not to prove that the founder was right. It is to help the venture become less wrong, one evidence-backed step at a time.