Use a semi-structured discussion guide split into introduction and consent, warm-up, topic exploration, an optional task or prototype walkthrough, and closing, with timing and probes built into each block. Map one research objective to three to five research questions before you write a single interview question. The rest of this guide gives you a copyable script template, question-writing techniques, and a pilot checklist so your sessions run on time and produce usable insight.
TL;DR:
- Your consent script should identify the researcher, purpose, data collected, sharing, retention, and withdrawal rights; capture remote verbal consent only after written consent was sent.
- Recheck consent before sensitive topics and recording changes, confirm retention limits and third party processing beforehand, and delete participant data if consent is withdrawn.
- Use open story prompts and neutral probes; replace hypothetical questions with observed tasks, then ask about specific friction instead of whether participants liked the experience.
- Pilot the guide with a colleague, time each block, and cut warm up first when sessions run short, never the consent script.
- After each session, separate direct observations from interpretations, draft a takeaway while details are fresh, and record three takeaways, notable quotes, and follow up actions.
Table of Contents
- A ready-to-copy user research interview script template
- How to write research questions and turn them into interview prompts
- Consent, recording, and data handling: what to say and how to prove it
- Moderation techniques that get real stories without leading the witness
- Pilot the guide, time each block, and run sessions on schedule
- Debrief fast: turning raw notes into findings the team can use
- Why the script is also a rehearsal document, not just a question list
- Rehearse your discussion guide with a voice-following teleprompter tool
- FAQ
- Sources
A ready-to-copy user research interview script template
Start your document with an agenda block for the team, not the participant: list the moderator, note-taker, observer, session length, and a timestamped running order. This keeps everyone aligned before the recording starts.
Introduction and consent (5 minutes) Read a short, plain-language script that covers who we are, why we’re talking to the participant, what we’ll do with the recording, and that they can stop at any time. A workable version: “Thanks for joining. We’re researching how people [complete a task]. There are no right or wrong answers, we just want to hear about your real experience. With your permission, we’ll record this session for note-taking purposes only. You can skip any question or stop at any point.” Confirm consent verbally and note the timestamp.
Warm-up (5 minutes) Ask easy, context-setting questions that build rapport before the real topic starts:
- “Tell me a bit about your role and a typical day.”
- “How long have you been doing [relevant task]?”
Topic exploration (20 to 30 minutes) Group your questions under each research question, with a broad starter and two or three probes underneath:
- Starter: “Walk me through the last time you [did the task].”
- Probe: “What happened right before that?”
- Probe: “What did you expect to happen next?”
Repeat this pattern for each research question so coverage is visible at a glance.
Optional task or prototype walkthrough (10 to 15 minutes) Give a neutral scenario instruction, such as “Imagine you need to [goal]. Show me how you’d try to do that here.” Stay quiet while the participant works, then probe what you observed rather than asking if they liked it.
Closing (5 minutes) Ask Gov directly: “What have I not asked that you think I should know?” and whether there’s anyone else or any other workflow worth studying. Explain next steps, confirm compensation, and do a final consent check before stopping the recording.
How to write research questions and turn them into interview prompts
Good interview questions start upstream, not in the interview document itself. Write your research objective first, then narrow it into three to five research questions that describe what you need to learn, not what you’ll ask. GOV.UK’s guidance on in-depth interviews recommends reviewing these questions, listing supporting topics, and ordering them logically before drafting a single participant-facing line.
From there:
- Brainstorm more candidate questions than you’ll use, then rewrite closed ones (“Do you like this?”) into open starters (“Tell me about a time you…”).
- Check each interview prompt against its research question and flag any objective with no matching question yet.
- Use the funnel technique: broad story prompts first, then clarifiers, then targeted probes that narrow toward specifics.
This mapping step is what turns a list of questions into a guide that produces evidence instead of opinions.
Consent, recording, and data handling: what to say and how to prove it
Your opening script needs to disclose who’s conducting the research, its purpose, what data you’re collecting, whether it’s shared, how long it’s kept, and that participation is voluntary with the right to withdraw at any point, according to consent guidance from User Research Education.
For evidence of consent:
- In person: a signed form kept with your research records.
- Remote: verbal confirmation captured on the recording, acceptable when the participant already received the written consent form beforehand.
- Either way: an emailed confirmation as a backup record.
Re-check consent before sensitive topics and before starting or stopping any recording or transcript. GOV.UK notes that consent is an ongoing process, and you should be ready to delete data if someone withdraws it. Before the study starts, confirm your retention limits and whether any third-party transcription or storage tool will process participant data.
Moderation techniques that get real stories without leading the witness
Open with story-starting prompts: “Tell me about the last time you [did the task],” then follow with “What happened next?” and “What did you expect?” NN/g’s research on question mistakes shows these neutral, concrete requests surface details a closed question never would.
Watch for these patterns and rephrase on the spot:
- Leading: “Didn’t you find that confusing?” becomes “How did that feel to use?”
- Compound: “Was it fast and easy?” becomes “Walk me through how that went.”
- Hypothetical: “Would you use this?” becomes an observed task instead, since NN/g notes hypothetical questions are weak evidence compared to watching actual behavior.
During a prototype segment, stay silent while the participant works, then probe what you saw rather than asking for a verdict.
Pro Tip: Let silence sit for a few extra seconds after a participant finishes talking. They’ll often add the most honest detail right after they think you’re done listening.
Pilot the guide, time each block, and run sessions on schedule
Pilot your script with a colleague before recruiting real participants. NN/g recommends this step specifically to catch confusing wording, awkward ordering, and timing that runs long.
- Time each block during the pilot and note where real conversation stretches past your estimate.
- Decide in advance what to cut first if you’re running short: usually the warm-up, never the consent script.
- Before the first real session, confirm roles, test your tech and recording setup, load any prototype links, and confirm how compensation will be delivered.
- Space real sessions far enough apart to write notes before the next participant arrives.
Debrief fast: turning raw notes into findings the team can use
Write down direct observations and exact quotes first, then separate your own interpretations into a clearly marked second column or section. This keeps evidence distinct from opinion when you write up findings later.
Right after each session:
- Note any follow-up questions or assumptions the session raised for future rounds.
- Draft a one-page takeaway while the conversation is still fresh.
- Capture your top three takeaways, the most notable quotes, and any follow-up actions.
This light structure turns a messy transcript into something your product team can actually act on within the hour.
Why the script is also a rehearsal document, not just a question list
A good interview script is read, re-read, and delivered out loud, so it needs the same care as any spoken document: short sentences, clear instructions, and cues for pacing. Moderators who rehearse the consent language and timing beforehand sound calmer and less scripted in the actual session. If you run remote interviews on video calls, a teleprompter-assisted rehearsal can help you internalize the opening and closing scripts so you’re reading the participant’s reactions, not your notes.
— D
Rehearse your discussion guide with a voice-following teleprompter tool
Reading a long discussion guide while holding eye contact is its own skill, and that’s exactly where our voice-following teleprompter earns its place in your prep routine. Instead of scrolling manually while you’re trying to listen, Voice Follow moves the script as you speak, so your consent language and transition lines stay on pace without breaking focus.
- Build chapters for each block of your guide, intro, warm-up, topic exploration, closing, and jump between them during practice runs.
- Save a playlist of guides if you’re running several studies with different scripts.
- Rehearse your exact consent wording until it sounds natural instead of read.
Try it at Pronoto before your next round of sessions.
FAQ
What is a user research interview script?
A user research interview script, or discussion guide, is a structured document that organizes introduction and consent, warm-up, topic exploration, an optional task walkthrough, and closing questions around your research objectives. It includes both the exact wording you’ll say to participants and moderator-facing notes like timing and probes.
How long should a user research interview take?
Most structured interviews typically run about an hour once you account for introduction and consent, warm-up, topic exploration, and closing. Piloting your specific guide is the only reliable way to confirm your actual timing, since topic depth varies by study.
How many questions should be in an interview script?
Rather than counting total questions, map three to five research questions to your study and build two or three starter and probe questions under each one. This keeps the guide focused on what you need to learn instead of padding it with extra items.
How do I get consent for a user research interview?
Your opening script should disclose who’s conducting the research, its purpose, what data you’ll collect, how it’s shared and retained, and the participant’s right to withdraw at any time, based on informed consent guidance. Capture evidence through a signed form, an emailed confirmation, or verbal consent recorded at the start of the session.
Should I ask participants if they would use a product?
Hypothetical questions like “would you use this?” are weak evidence on their own, since NN/g’s research on interviewing users shows observed behavior during a task is far more reliable. Give the participant a neutral scenario, watch them work, and probe the friction you actually see.
Sources
- Gov
- 6 Mistakes When Crafting Interview Questions - NN/g
- Getting informed consent for user research - User Research Education guidance