A user interview is a research method where you ask participants questions about a topic, listen to their answers, and follow up to learn more. This guide covers how to plan one, which questions to ask, how to record and transcribe it, and how to analyze what you learned. It includes four free Word templates: an interview guide, a recording consent script, a note sheet, and a transcript format, each with a filled example. If you would rather not take notes during the interview at all, skip to how Comu AI turns the recording into notes.
The same steps apply to customer interviews, research interviews, and journalist interviews. The templates are written for user research, and you can change the wording for other uses.
What is a user interview?
The Nielsen Norman Group describes a user interview as a research method where the interviewer asks participants questions about a topic, listens to their responses, and follows up with further questions to learn more. It adds that the term "user interview" is unique to the UX field. In market research or social science, the same method is called an in-depth interview, a semi-structured interview, or a qualitative interview.
User interviews are used in the discovery phase, to learn what people's experiences are like, what they find memorable, where they struggle, and what they care about. A customer interview is the same method with customers as the participants.
Free user interview templates (Word downloads)
Each file has two pages. Page one is blank. Page two is the same template filled in with made-up details, so you can see how much to write.
| Template | Use it to | Download |
|---|---|---|
| User interview guide | Plan the introduction, warm-up, main questions, and wrap-up | Word file |
| Interview recording consent | Tell the participant what you record, how it is transcribed, and their choices | Word file |
| Interview note sheet | Capture time, quotes, and takeaways while you listen | Word file |
| Interview transcript format | Lay out a transcript with time, speaker, and text | Word file |
All four are plain Word tables with no macros. Short on time? Let Comu AI turn the recording into notes. They work in Google Docs too: upload the file to Drive and open it with Docs. The consent template is a starting point, not legal advice, so check your own review board or employer rules.
How to conduct a user interview in six steps
The steps below follow the Nielsen Norman Group's guide to user interviews.
- Identify what you would like to learn. Write down the research question before you write any interview questions.
- Prepare a guide. The guide directs the conversation. It should contain a few well-designed, open-ended questions that get participants talking and sharing their experiences.
- Pilot the guide. Try it with a colleague or a target user first. Piloting shows you which questions were misunderstood, which to add or remove, and whether the order felt natural.
- Start easy. Explain the purpose of the interview, the kinds of questions you will ask, and how the information will be used. Start with questions that are easy to answer.
- Build rapport. People talk more when they feel relaxed and trust the process. Show you are listening with verbal and nonverbal cues, and ask related questions.
- Follow up and probe. Ask your prepared follow-up questions if the participant did not cover them, and ask further questions that dig into their answers.
The interview guide template gives you a place for each part: introduction, warm-up, main questions with probes, and wrap-up.
User interview questions
Good questions are open-ended and ask about specific events, not general habits. The Nielsen Norman Group gives these examples of good open-ended questions:
- Walk me through a typical day for you.
- Tell me about the last time you [did something].
- Tell me about a time when you [had a particular experience].
It advises jogging the memory by asking about specific events rather than general processes. It also lists probing questions, such as "Tell me more about that," "Can you expand on that?" and "Why is that important to you?" You do not usually prepare probes in advance, but if you might forget them, write them at the top of your guide or on a card.
A short user interview guide example
Here is the example on page two of the guide template, in plain text:
| Part | Content |
|---|---|
| Introduction | Thank the participant, explain the purpose, and ask if you may record |
| Warm-up | Tell me a bit about yourself and who you cook for |
| Main question 1 | Walk me through how you planned meals last week. Probe: tell me more about that |
| Main question 2 | Tell me about the last time a plan fell apart. Probe: what did you do instead? |
| Wrap-up | Is there anything I did not ask that you would like to add? |
The study, questions, and people in the example are made up.
Customer interview template and questions
A customer interview uses the same guide. Change the goal and the warm-up: the goal is usually to learn why customers chose a product, how they use it, and what is missing, and the warm-up asks about their role and how they work. Keep the main questions about specific events, such as the last time they used the product or the time they decided to switch, in line with the Nielsen Norman Group's advice to ask about specific events.
If you interview customers, tell them at the start what the interview is for and how the answers will be used. That is the same advice the Nielsen Norman Group gives for any participant.
How to record an interview
Ask for permission before you record, and tell the participant what will happen to the recording. A consent form for audio-recorded interviews from Quincy College gives a clear model. It says the interview will be audio recorded, that neither the participant's name nor other identifying information will be associated with the recording or the transcript, that only the research team will listen, and that the recording will be erased once the transcript is checked for accuracy. It also gives the participant the chance to have the recording erased right after the interview if they withdraw consent.
Our consent template follows the same structure, with one addition: it asks you to say who or what will transcribe the recording, such as a person, software, or a service. If a third party handles the audio, say so in the form.
Practical tips for recording:
- Test first. Record ten seconds and play it back before the participant arrives or joins.
- Place the recorder close. For an in-person interview, put it on the table between you and the participant, not in a bag or pocket.
- Reduce noise. Choose a quiet room, and turn off fans and notifications.
- Name the file. Use the date and the participant code, not the name, such as `interview-p4.mp3`.
- Back it up. Copy the file to a second place as soon as you finish.
Tell the other person you are recording. Laws on recording calls differ by country, and you are responsible for following them. This is general information, not legal advice.
How to transcribe an interview
A transcript turns the recording into text you can search, quote, and analyze. The steps are the same whether you type it yourself or use software. How to transcribe audio to text compares five ways to get a transcript, with the limits of each.
- Choose a style. Verbatim keeps every word, including filler words and false starts. Clean verbatim removes them. Pick one and use it throughout. Our transcript template has a field for the style.
- Label the speakers. Use "Interviewer" and a participant code such as P4, not the participant's name.
- Add timestamps. A time at each change of speaker makes it easy to find a quote in the recording later.
- Mark unclear parts. Write [inaudible] and the time, instead of guessing.
- Check against the recording. Listen to the recording while you read the transcript. This is the step that catches wrong names, numbers, and terms.
- Remove identifying details. Replace names, employers, and places that could identify the participant.
The Quincy College form describes this same order: the researcher transcribes the interview and the recording is erased once the transcript is checked for accuracy. If you use software to draft the transcript, the check against the recording matters even more. If checking by hand is the part you dread, see how Comu AI drafts the notes from the recording.
User interview analysis: a simple method
Analysis is where interviews become useful. The steps below are our own suggested method, not taken from a source. Researchers use many approaches, and a review board or your team may require a specific one.
- Read each transcript once without marking anything. Get a sense of the whole conversation.
- Highlight quotes that answer your research question. Pull the participant's words, not your interpretation.
- Group similar quotes. Put them on cards or in a table, and name each group.
- Count how many participants mention each group. A theme several participants raise deserves more weight than one only one person mentioned.
- Write the takeaways. Three to five sentences, each backed by quotes, and note what you did not learn.
The note sheet template supports this: it has columns for time, what the participant said, the quote, and your note, and a place for the top three takeaways.
In person, phone, or video: how the format changes the recording
The method is the same in every format, but the recording is not. Pick the format first, then plan the recording around it.
| Format | Record with | Watch out for |
|---|---|---|
| In person | A recorder on the table between you and the participant | Room noise, and a recorder placed too far away |
| Phone | A recorder or app that can capture both sides of the call | Only recording your own voice, and consent rules for calls |
| Video call | The call software's own recording, or a separate recorder as backup | Participants who join with a bad connection, and a recording that fails silently |
Whatever the format, record a backup if you can. A second recorder, even a phone voice memo, costs nothing and saves the interview if the first one fails. For a call, say aloud at the start that you are recording, so it is on the recording itself.
For phone and video interviews, check the consent rules that apply to the people on the call. The rules differ by place, and you are responsible for following them.
How long should a user interview be, and how many should you run?
There is no single answer, and we do not cite a number here. The Nielsen Norman Group notes that interviews are a qualitative method, so it is fine to keep making minor tweaks to the guide as you learn. A practical approach is to set the length from your guide, since three to five main questions fit in about half an hour, and to keep running interviews until new ones stop telling you anything new. When the same points keep coming up and nothing new appears, you have enough for this question.
If the topic has several kinds of users, interview each kind, since one group's answers may not apply to another.
Interview questions template: how to structure the list
An interview questions template is the interview guide with the questions written for your topic. The guide template above gives you the frame, and this is how to fill it in.
| Part | What to write | Example |
|---|---|---|
| Opening | Purpose, length, how answers are used, recording question | "This takes about 30 minutes. May I record it for my notes?" |
| Easy start | One or two questions the participant can answer without thinking | "Tell me a bit about yourself." |
| Main questions | Three to five open-ended questions about specific events | "Tell me about the last time you [did something]." |
| Probes | Short follow-ups, written beside each main question | "Tell me more about that." |
| Close | An invitation to add anything, and a referral question | "Who else should I talk to?" |
Keep the list short. Three to five main questions is plenty for a 30 minute interview, since the follow-ups take the time. Write the questions so that a participant cannot answer with one word. Put the question you most need answered in the middle, after the participant has warmed up and before they tire.
Read the list aloud once before the first interview. Questions that look fine on paper often sound leading when spoken.
Interview transcription software: what to check
If you use software to draft the transcript, judge it on your own recording, not on a feature list. Record five minutes of a real interview, run it through each tool you are considering, and compare the output against the audio. These are the points to check.
- Accuracy on your audio. Check names, numbers, and technical terms, since those matter most in a quote.
- Speaker labels. An interview has at least two voices, so the tool should separate them. Check whether the labels stay correct when people interrupt.
- Timestamps. You will want to find the quote in the recording later.
- Editing. You will correct the draft, so editing should be easy.
- Export. Check that you can export to Word, text, or PDF, so you can move the transcript into your analysis.
- Where the audio goes. For research, your consent form should say whether a third party processes the recording. Check each tool's own privacy statement.
- Cost. Compare the price for the hours you actually record, and check what happens when a free allowance runs out.
Before, during, and after the interview: a checklist
| When | Do this |
|---|---|
| Before | Write the research question. Draft the guide from the template. Pilot it. Prepare the consent form. Test the recorder. |
| At the start | Explain the purpose and how answers are used. Ask for consent to record. Start recording and say the date and participant code aloud. |
| During | Ask open-ended questions. Probe. Note timestamps of key moments on the note sheet. Do not interrupt. |
| At the end | Ask if there is anything to add. Thank the participant. Stop the recording. |
| Within a day | Back up the file. Transcribe or draft the transcript. Check it against the recording. Remove identifying details. |
| After all interviews | Group quotes, count how many participants mention each group, and write the takeaways. |
Common user interview mistakes
- Asking leading questions. A question like "Don't you find it hard to plan meals?" tells the participant the answer you expect. Ask "How do you plan meals?" instead.
- Talking too much. Aim to listen most of the time. If you are speaking more than the participant, the guide has too many questions or too little silence.
- Skipping the pilot. The first interview is when you find out which questions do not work. A pilot moves that discovery before the real participants.
- Recording without saying so. Always ask first, and put it in writing for research.
- Transcribing without checking. A transcript you have not checked against the audio can put wrong words in a participant's mouth.
- Waiting too long to write up. Notes and impressions fade. Write the takeaways within a day of each interview.
Let Comu AI turn your interview recording into notes
An interview asks you to listen closely, think of the next question, and take notes, all at once. A recording removes the last task.
Comu takes out the manual step. Its built-in summary templates already follow structures like the ones above. Record the interview with a Comu recorder, and the app transcribes live, names the speakers, and turns the recording into a finished note in the template you choose. You do not type the notes or build the template yourself. Actions pulls out tasks and decisions, and you can edit the summary afterward. There are more than 30 built-in templates, and you can create as many of your own as you like. All of this is included on the free plan that comes with the device. Export the notes as TXT, Markdown, Word, or PDF. AI-generated results are meant to assist you and may need review and adjustment before final use.
For research interviews, tell participants in the consent form that the recording will be transcribed by software, and check the transcript against the audio before you quote it. If your review board has rules about third-party tools, check them first.
Comu Note Pro is thin enough to slip into a notebook, at 3.1 mm. Comu Action Pro has six microphones and about 7 m of indoor pickup, which suits a group interview or a focus group. For the wider set of features, read our guide to the features that come with Comu.
Common questions about user interviews
What is a user interview?
A user interview is a research method where the interviewer asks participants questions about a topic, listens to the answers, and follows up to learn more. The Nielsen Norman Group says the term is unique to UX. Other fields call the same method an in-depth, semi-structured, or qualitative interview.
How do you conduct a user interview?
Decide what you want to learn, write a guide of a few open-ended questions, pilot it, start with easy questions, build rapport, and follow up with probes. The six steps above walk through this, and the interview guide template is set up for it.
What questions should you ask in a user interview?
Ask open-ended questions about specific past events, such as "Tell me about the last time you did X" or "Walk me through a typical day." Follow up with probes like "Tell me more about that."
How do you record an interview?
Ask for permission and explain what happens to the recording. Test the recorder first, place it close to the participant, and use a quiet room. Name the file with a participant code and back it up. The recording consent template helps you cover the points a participant should know.
How do you transcribe an interview?
Choose verbatim or clean verbatim, label the speakers, add timestamps, mark unclear parts, and check the transcript against the recording. You can type it yourself or draft it with software and then check it. The transcript template shows the layout.
How do you analyze user interviews?
Read each transcript once, highlight quotes that answer your research question, group similar quotes, and count how many participants mention each group. Then write three to five takeaways, each backed by quotes.
Do you need consent to record an interview?
Ask for it, and say so in writing for research. Laws differ by country, and research projects usually have review board rules. The consent template is a starting point, not legal advice.
Can AI transcribe interviews?
Yes. Software can draft a transcript from a recording. It can mishear names, numbers, and technical terms, so check the draft against the audio. For research, tell participants that software will transcribe the recording.
Related reading
- Meeting Minutes Template: Free Word Downloads and Examples
- Cornell Notes and Note-Taking Methods: Free Templates and Examples
Sources
Pages were checked on Oct 31, 2026.
- Nielsen Norman Group, User Interviews 101
- Quincy College, Consent Form for Audio Taping and Transcribing Interviews
- Comu, Comu Action Pro and Comu Note Pro
