Skip to main content

UX and research · Beginner · 14 min

Write an interview guide without leading questions

By Padding · Updated 18 Nov 2026
Any document editor · November 2026

You will have a 45-minute interview guide built from your research questions, with a warm-up, story-based questions, follow-up prompts and a close, plus a list of leading questions you caught and rewrote.

On this page

An interview guide is the sheet you hold during a user interview. It is not a script you read word for word. It is a running order: what to cover, roughly how long to spend on each part, and the exact wording of the questions that matter most, so you do not improvise them badly under pressure.

The most common fault in a beginner's guide is the leading question. "Wouldn't it be easier to book online?" tells the person what you hope to hear, and most people are polite enough to say it back. The answer feels like evidence and is worth nothing. A good guide removes those questions before anyone is in the room.

This tutorial uses a running example: Larkfield Clinic, a fictional community clinic planning an appointment booking app where patients choose a service, pick a time, confirm and get a reminder. You are interviewing patients about how they book appointments today. Swap in your own product as you go.

By the end you will have one document with two parts: a 45-minute guide, and a "caught and rewritten" list showing each leading question you wrote first and the version that replaced it. Keep that list. It is the quickest way to train your ear for the next guide.

Step 1 · Separate research questions from interview questions

Open a new document and title it with the study name and date, for example "Larkfield booking interviews, guide v1". Paste your research questions at the top under a heading "What we need to learn".

Research questions are what you want to know. They are often abstract, and many of them are leading if asked directly. For Larkfield they might be:

  • How do patients book appointments today, and where does it go wrong?
  • What makes a patient miss or cancel an appointment?
  • Would patients trust an app to hold their booking?

You do not ask any of these out loud. "Would you trust an app with your booking?" invites a guess about the future, and people are poor at predicting their own behaviour. Instead, each research question becomes a topic you will ask about through the person's past experience.

Under each research question, write one line naming the experience that would answer it:

  • How patients book today: the last appointment they booked, from start to finish.
  • Missed or cancelled appointments: the last time they missed, moved or nearly missed one.
  • Trust: what they did the last time they booked something important online, and how they checked it worked.

Check it worked

Every research question at the top of your document has one line under it starting with "the last time" or naming a specific past event. None of those lines asks the participant for an opinion or a prediction.

Step 2 · Block out 45 minutes

Before writing a single question, decide how the time is spent. Without a budget, the first topic eats the session. Add this outline below your research questions, adjusting the minutes to your own topics:

  1. Welcome and consent (3 minutes)
  2. Warm-up (5 minutes)
  3. Story 1: the last booking (15 minutes)
  4. Story 2: the last missed or moved appointment (10 minutes)
  5. Story 3: trusting an online booking (7 minutes)
  6. Close (5 minutes)

That adds up to 45. Put your most important topic first among the stories, while the person is fresh and you have the most time. If something has to be cut on the day, it should be the last story, so order them by priority.

Write the clock time each section should start beside it (0:00, 0:03, 0:08, 0:23, 0:33, 0:40). During the interview a glance at the clock tells you whether to move on.

Check it worked

Your outline has six sections whose minutes add up to 45, each with a start time, and the story you care about most comes first.

Step 3 · Write the welcome and warm-up

The welcome sets expectations and covers consent. Write it out in full, because you will say it at the start of every session and it should be the same each time. For Larkfield:

Thanks for making time. I'm [name] and I work on a project for Larkfield Clinic about how people book appointments. There are no right or wrong answers, and I'm not testing you. I didn't design anything you'll mention, so you won't hurt my feelings. With your permission I'd like to record this so I can listen back; the recording stays with the research team. Is that all right? You can stop at any point.

Only make the "I didn't design anything" promise if it is true. If you did design something, say instead that honest criticism is the most useful thing they can give you.

The warm-up gets the person talking about something easy and related, so the first real question is not their first sentence. Write two or three questions:

  • Tell me a bit about yourself and what a usual week looks like.
  • How often do you end up needing a doctor or a clinic, roughly?
  • Who else in your household do you book appointments for, if anyone?

The last one is a warm-up that also earns its place: someone who books for a parent and two children has a very different booking life from someone who books once a year for themselves.

Check it worked

Your welcome mentions recording, consent and the right to stop, and your warm-up has two or three questions that a stranger could answer comfortably in under a minute each.

Step 4 · Write story questions about the last time

The core of the guide is a few questions that ask for a specific, recent story. A specific story is harder to embellish than a general habit. "How do you usually book?" gets the idealised version. "Tell me about the last time you booked" gets what happened, including the bits that went wrong.

For each story section, write one opening question that names a real event and asks the person to walk you through it:

  • Story 1: Think back to the last appointment you booked, for you or anyone else. Take me through it from the moment you realised you needed one.
  • Story 2: Can you remember the last time you missed an appointment, moved one, or nearly forgot one? What happened?
  • Story 3: Think of the last time you booked something that mattered to you online, such as a train, a payment or an appointment. How did you know it had gone through?

Notice that Story 3 does not mention apps, clinics or trust. It asks about a past behaviour and lets the person show you what reassures them. That is your trust research question, answered without asking it.

Under each opening question, list the moments you hope the story covers so you can steer gently if they skip one. For Story 1: deciding to book, choosing which service, finding a time that worked, giving their details, knowing it was confirmed, being reminded. These match the Larkfield app's screens (Home, Choose a service, Pick a time, Your details, Confirm, Booked), but do not read them out. They are a checklist for you.

Check it worked

Each story section has exactly one opening question containing a past-tense event ("the last time", "can you remember"), followed by a checklist of moments that is clearly marked as notes for the interviewer, not questions.

Step 5 · Catch and rewrite your leading questions

Now go through every question you have written, plus any extra ones you were tempted to add, and test each one. A question is leading if any of these is true:

  • It suggests an answer: "Wouldn't it be easier to...", "Don't you find...".
  • It contains your opinion or a judgement word: "frustrating", "confusing", "simple".
  • It asks about the future or a hypothetical: "Would you use...", "How likely are you to...".
  • It offers only yes or no where you want a description.
  • It asks two things at once, so the person answers the easier half.

Add a heading "Caught and rewritten" at the bottom of your document. For every question that fails, write the original, the problem, and the replacement. Here are the ones beginners most often write for a booking study, with rewrites:

  • Before: "Isn't phoning the clinic frustrating?" Problem: puts the word "frustrating" in their mouth. After: "How did you feel about phoning the clinic that time?"
  • Before: "Would you use an app to book appointments?" Problem: a prediction about the future; almost everyone says yes. After: "What did you use to book the last time? What made you pick that?"
  • Before: "Do you forget appointments?" Problem: yes or no, and embarrassing to admit. After: "What do you do to remember an appointment once it's booked?"
  • Before: "How useful would a reminder text be?" Problem: assumes reminders are the answer and asks them to rate your idea. After: "The last time you nearly missed something, what happened just before?"
  • Before: "Did you find a time easily, and was the receptionist helpful?" Problem: two questions; they will answer one. After: "How did you end up with the time you got?" Then, separately, ask about the receptionist only if they mention them.
  • Before: "Our new app lets you pick a slot in three taps. What do you think?" Problem: pitches a solution and asks for approval. After: Remove it. This is a usability test question, not an interview question.

Then replace the originals in the guide itself with the "after" versions.

Check it worked

Your "Caught and rewritten" list has at least four entries, each with a before, a problem and an after, and a search of the guide part of your document finds none of the words "wouldn't", "don't you", "would you", "frustrating", "easy" or "simple".

Step 6 · Add follow-up prompts

The opening question starts the story. Follow-up prompts keep it going and get underneath it. Write a short bank of neutral prompts at the top of the guide, where you can see them at any point:

  • What happened next?
  • Can you tell me more about that?
  • Why was that?
  • What did you do then?
  • You said [their word]. What do you mean by that?
  • How did that compare with other times?
  • (Silence. Count to five before you speak.)

These work on any story because they carry no content of their own. The prompt that repeats their word back is the most useful: if someone says booking was "a faff", ask what made it a faff, using their word, not yours.

Under each story section, add one or two prompts specific to that story that you will use only if the person does not cover the topic themselves. For Story 2: "How did you find out you'd missed it?" and "What happened with the clinic afterwards?" Keep them neutral; run them through the leading-question test from the previous step too.

Check it worked

The guide has a general prompt bank of five to eight lines near the top, and every story section has at most two specific prompts, each of which passes the leading-question test.

Step 7 · Write the close

The last five minutes catch what you missed and leave the person feeling the time was worth it. Write three things:

  1. A catch-all: "Is there anything about booking appointments we haven't talked about that you think I should know?" This often brings out the most surprising answer of the session.
  2. A wrap-up: thank them, say what happens next (for example, "we're talking to eight people this month, and what you've said will shape what the clinic builds"), and explain any thank-you payment or voucher and when it arrives.
  3. A permission question, if you want it: "Would you be happy for us to contact you again if we test an early version?" Note their answer.

Do not finish with "So, would you use our app?". It undoes the work of the whole session in one sentence.

Check it worked

Your close section contains a catch-all question, a thank-you with what happens next, and no question about your product or idea.

Step 8 · Read it aloud and trim to fit

Read the whole guide aloud at a normal speaking pace, answering each question yourself in a sentence or two. Time it. If your read-through runs over 25 minutes, the real interview will run well past 45, because participants talk more than you do in a read-through and the follow-ups add time.

Cut in this order until it fits:

  1. Warm-up questions beyond two.
  2. Specific prompts the general bank already covers.
  3. The lowest-priority story (the last one).

Then save the guide as a new version ("guide v2") and, if you can, try it with one colleague or friend who is not on the project. Note any question they asked you to repeat or answered with "it depends". Those are unclear, and need rewording before the first real participant.

Check it worked

Your document has a v2 that reads aloud in 25 minutes or less, its section minutes still add up to 45, and your "Caught and rewritten" list is still at the bottom with any extra rewrites from the read-through added.

Next steps

Take the guide into a session with Run your first user interview, which covers booking participants, running the conversation and taking notes. If you do not yet have research questions tied to a decision, start with Write a one-page research plan. If your research questions came from the people funding the work, Interview stakeholders at the start of a project shows how to gather and compare them first.

Need this done for you?

Padding does this work for growing companies, from the first diagnosis to a system that keeps working.

Digital Transformation

Get new tutorials

New tutorials, about twice a month.