Recruit customer interview participants by defining the experience you need, writing a short neutral screener, using several appropriate channels and inviting people without pressuring them. Explain the purpose, format, data use, recording and incentive before the session. The goal is not a convenient group of supporters; it is a relevant group that can provide evidence for a specific decision.
Start with evidence, not demographics
Translate the research question into observable recruitment criteria. If you need to learn why startups replace spreadsheets with planning software, relevant participants might have evaluated or adopted such a tool in the last six months. “Works in technology” is too broad.
Create three lists:
- Must have: experience essential to answering the question.
- Useful variation: role, company stage, outcome or alternative that may change the account.
- Exclude: conflicts or circumstances that would make the session unsuitable.
Avoid criteria that merely reproduce the team’s assumptions. Include different access needs where the product has a broad audience. GOV.UK guidance says effective participants should be actual or likely users and recommends including the different kinds of people who may need the service.
Write a neutral screener
A screener should confirm eligibility without revealing the answer you hope to hear. Ask about recent behaviour before opinions.
For example:
- What is your role in decisions about [relevant task]?
- When did your organisation last change how it handles that task?
- Which approaches did you consider?
- Which option did you choose, if any?
- Were you personally involved in evaluation or use?
- Do you have any access requirements for a remote session?
Do not publish the exact qualifying combination if it would encourage applicants to game the screener. Collect only information you need. ICO guidance describes data minimisation as limiting personal data to what is necessary, and storage limitation as keeping it only as long as needed.
Use more than one recruitment channel
Suitable channels can include:
- recent customers, losses or former customers in your records;
- product intercepts or research invitations sent through appropriate systems;
- customer support or account teams, using a consistent brief;
- professional communities where participation requests are allowed;
- specialist recruitment agencies for difficult or sensitive audiences;
- partner organisations with access to the relevant group.
Founder networks are fast but often over-represent people who resemble the founder. Customer-success nominations may favour satisfied, articulate accounts. Paid panels can increase reach but still require careful screening. Record the channel next to each participant so you can see where bias may have entered.
Use how many customer interviews you need to structure rounds rather than filling an arbitrary quota.
Make the invitation clear and voluntary
A useful invitation states:
- who is conducting the research;
- what broad subject will be discussed;
- why the person was invited;
- session length and format;
- whether recording is requested;
- the incentive and payment timing, if offered;
- how to decline or request adjustments;
- what happens to the data.
Do not make continued access, customer support or a commercial relationship feel conditional on participation. Send an information sheet and consent process in advance. GOV.UK guidance says informed consent should cover the purpose, data collected, use, sharing, retention, recording and the right to stop or withdraw.
Illustrative example
We are researching how operations leaders evaluate reporting workflows. We would like to discuss a recent evaluation, whether or not you selected our product. The 45-minute video call is research, not a sales meeting. With your permission, we will record it for analysis by the research team and delete the recording under our stated retention policy. Participation is voluntary and you may stop at any time.
This invitation is an illustrative template. Adapt it to the actual controller, data use, lawful process and incentive; do not paste it as a complete privacy notice.
Protect quality on the day
Confirm consent again, check recording permission and give participants a chance to ask questions. If observers are present, say who they are. Pay promised incentives promptly and do not make payment conditional on praising the product or completing every question.
Recruit replacements for no-shows in advance rather than accepting unsuitable participants. Keep contact details separate from research notes where possible. Give participant IDs to notes and transcripts, and restrict access to people who need it.
For the session itself, use neutral prompts from customer interview questions for better positioning.
Limitations and trade-offs
Recruiting only existing customers is operationally easy but misses the status quo, lost deals and people who rejected the category. Recruiting strangers can reveal alternatives but may require more context and screening. Incentives improve participation and acknowledge time, but they do not remove selection bias.
No recruitment method creates a statistically representative sample by itself. Describe who participated, how they were found and who was absent. If a claim concerns how common something is, test it with an appropriate quantitative method.
Frequently asked questions
Should founders recruit customers personally?
They can, especially at an early stage, but the relationship may make customers overly polite. Use neutral wording, include people who did not buy, and consider having someone else conduct sensitive interviews.
Should we pay interview participants?
Often, yes: an incentive recognises time and can reduce barriers. Set it according to session burden and audience norms. State the amount and terms clearly, and never tie payment to a favourable response.
Can salespeople select participants?
Sales teams can help identify relevant people, but should use documented criteria rather than choosing only strong relationships. Balance their nominations with other channels and make clear that the interview is not a sales call.
Your useful next step
Draft a one-page recruitment brief containing the decision, must-have criteria, useful variation, exclusions, channels, incentive and consent owner. Ask a colleague to try the screener without seeing your intended answers. After the round, record recruitment limits alongside the findings and compare the result with interviews versus surveys.