Skip to content

afeezadebayo/research-learning-skills / expert-user-interviews

expert-user-interviews

Verified

Plan and run customer, user, buyer, churn and expert interviews: learning goals, recruiting by behaviour, discussion guides with past-behaviour questions, good versus leading questions, consent and recording, note-taking, debriefs, expert interview tips and handover to synthesis. Use whenever someone needs to interview customers, users, prospects or experts, write an interview guide, run win-loss or churn interviews, recruit participants or avoid leading questions in research.

Expert and User Interviews

Some answers only exist in people's heads: why customers switched, how a buyer really decides, what an industry expert expects next year, where a process breaks. Interviews uncover them, but poorly run interviews mislead. Leading questions invite agreement, hypothetical questions ("would you use this?") produce polite guesses, and talking to the wrong people produces confident but irrelevant answers. Rob Fitzpatrick's The Mom Test summarises a practical rule: ask about people's past behaviour and specific experiences, not their opinions about your idea. [Reference, Fitzpatrick, The Mom Test]

This skill plans and runs interviews with customers, users, prospects and experts: defining learning goals, recruiting the right people, writing discussion guides with open, behaviour-based questions, running and recording interviews ethically, taking useful notes, avoiding bias and turning conversations into evidence.

Operating principles

  1. Learning goals first. Know what decisions the interviews inform.
  2. Talk to the right people. Recruit by behaviour and role, not convenience.
  3. Ask about the past, not the hypothetical. "Tell me about the last time..." beats "Would you...?"
  4. Listen more than you talk. Aim for the interviewee speaking most of the time.
  5. Follow the story. Ask why, how and what happened next; dig into specifics.
  6. Respect consent and privacy. Explain the purpose, recording and use of data.

Interview types

TypePurposeTypical number
Customer discoveryUnderstand problems and current behaviour8 to 15 per segment
Win and loss interviewsWhy buyers chose or rejected you10 to 20 per quarter
Churn interviewsWhy customers left8 to 12
Usability or task interviewsHow people use a product5 to 8 per round
Expert interviewsIndustry knowledge, forecasts, context3 to 10
Stakeholder interviewsInternal needs and constraintsAll key stakeholders
Patterns often start repeating after about 8 to 12 interviews in a well-defined segment; keep going if new themes still appear.

Good and bad questions

AvoidAsk instead
Would you use a tool that does X?Tell me about the last time you had to do X.
Do you like our product?Walk me through how you used it last week.
How much would you pay?What do you currently spend on solving this?
Isn't onboarding confusing?What happened when you first set it up?
Why didn't you buy? (asked bluntly)How did you make the decision? Who else was involved?

When to use, and when not to

Use it for customer research, product discovery, win and loss analysis, churn reasons, buyer journeys, expert input for market research and internal stakeholder research.

Don't use interviews to measure how common something is (use surveys or data) or to sell disguised as research.

Inputs

  • Learning goals and the decision they inform.
  • Target interviewees by segment, role and behaviour.
  • Access: customer lists, networks, recruiting panels, expert networks.
  • Incentives, consent forms and recording tools.
  • Timeline.

Workflow

Step 1. Define learning goals

Three to five things you need to learn, tied to decisions.

Step 2. Define and recruit participants

Screening criteria by behaviour (for example "switched accounting software in the last 6 months"); recruit across the segments that matter; offer appropriate incentives.

Step 3. Write the discussion guide

Warm-up, context, the main story questions, probes, wrap-up. Keep it to 8 to 12 core questions for 30 to 45 minutes.

Step 4. Prepare logistics

Consent, recording permission, note-taker or transcription, quiet setting.

Step 5. Run the interview

Explain purpose, start broad, follow specifics, avoid pitching, use silence and "tell me more".

Step 6. Capture notes immediately

Key quotes, behaviours, problems, workarounds, triggers, surprises; tag by learning goal.

Step 7. Debrief

After each interview, write a five-line summary; after every few, look for patterns and adjust the guide.

Step 8. Synthesise

Across interviews, group findings into themes with evidence (see research-synthesis).

Discussion guide template

Intro (2 min): purpose, consent, recording, no right or wrong answers
Context (5 min): role, team, responsibilities, tools used
Story (20 min):
- Tell me about the last time you <did the task / faced the problem>.
- What triggered it? What did you do first? Then what?
- What was hardest? Why?
- What did you try instead? How did that go?
- Who else was involved? How was the decision made?
Current solution (5 min): what you use now, cost, what you would change
Wrap-up (3 min): anything we did not ask? Who else should we talk to?

Expert interview tips

  • Prepare with desk research so you ask what only they can answer.
  • Ask for reasoning, not only predictions: "What makes you think that?"
  • Ask about uncertainties and what would change their view.
  • Respect confidentiality and compliance rules, especially through expert networks.

Probing techniques

TechniqueExample
Ask for specifics"Can you remember the exact moment that happened?"
Ask why, gently"What made that important?"
MirrorRepeat their last words as a question to invite more
SilenceWait three seconds after an answer; people often add the real reason
Ask for evidence"Could you show me how you do that?"

Note-taking template

Interviewee: <role, segment, company size> | Date | Interviewer
Context: <...>
Key story: <trigger, steps, outcome>
Problems and workarounds: <...>
Quotes: "<verbatim>" (time stamp)
Surprises: <...>
Tags: <learning goals>

Recruiting messages

Keep invitations short: why you are reaching out, what the conversation is about (not a sales call), how long it takes, the incentive and how to book. Recruiting from churned customers, lost deals and non-customers gives a truer picture than talking only to fans.

Using AI with interviews

Transcription and AI summaries save time, but check summaries against transcripts for key quotes and decisions, remove personal data before sharing widely and store recordings according to consent and retention rules.

Thank every participant and share what changed.

Edge cases and judgement calls

SituationWhat to do
Interviewee keeps giving opinionsRedirect to a specific past example
Interviewee is a big customerSeparate research from sales; involve the account owner
Hard-to-reach rolesUse referrals, LinkedIn outreach, incentives and short slots
Sensitive topicsExtra care with consent, anonymity and data storage
Different languagesNative-speaking interviewers or interpreters; translate notes carefully
Team members want to pitchAgree roles beforehand: research only

Handoffs to related skills

  • Planning: research-planning
  • Surveys to quantify findings: survey-design
  • Synthesis: research-synthesis
  • Meeting notes: the Meetings and Calendar meeting-notes-actions skill
  • Persuasion of stakeholders afterwards: the Communication and Presentations persuasion-influence skill

Worked example

A SaaS team believed customers churned because of price. Twelve churn interviews asked about the last weeks before cancelling: what happened, who decided and what they moved to. Most stories described failed onboarding and a champion leaving, with price mentioned only as the final reason given to finance. The team changed its retention focus to onboarding and champion handovers, and churn fell over the next two quarters.

Common mistakes

  • Asking hypothetical questions about the future.
  • Pitching instead of listening.
  • Interviewing whoever is easiest to reach.
  • Taking poor notes and relying on memory.
  • Treating one strong opinion as a pattern.

Output format

1. Learning goals
2. Participant criteria and recruitment plan
3. Discussion guide
4. Consent and logistics checklist
5. Interview summaries
6. Emerging themes for synthesis

Self-check before you answer

  • Are learning goals tied to decisions?
  • Are participants recruited by behaviour and role?
  • Do questions focus on past behaviour, not hypotheticals?
  • Are consent, notes and summaries handled properly?

Created by Afeez Adebayo, Marketing, SEO & Growth Expert Website: https://afeezadebayo.com LinkedIn: https://www.linkedin.com/in/afeez-adebayo/ X: https://x.com/za_icon Source: https://afeezadebayo.com/skills-md/productivity/research-learning/expert-user-interviews/

Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). You may copy, share and adapt this skill, including for commercial use, as long as you credit Afeez Adebayo and link to the source URL above. Keep this notice in any copy or adapted version.

add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks.Stars: 115kUpdated Claude CodeCodexVerifiedInstall