How to ask follow-up questions in user interviews without leading the participant

A good follow-up question should do one job: help the participant say what they mean. It should not smuggle in your hypothesis. That is not a small style issue.

What leading follow-ups do wrong

Leading probes often sound helpful because they are specific. They are not. A question like “Was it hard to find the pricing page?” assumes the page was hard to find. If the participant says yes, you get agreement, not evidence. If they say no, you may still have planted the idea that “hard to find” is the frame that matters.

That can push a participant toward a version of the story that was not theirs to begin with.

There is another failure mode that gets missed. Leading questions can overstate confidence. When you ask, “So the main issue was speed, right?” you are not just suggesting an answer. You are asking the participant to compress a mixed, uncertain experience into a neat label. That kind of prompt can turn a vague memory into a crisp claim I would not want to assume the person would have made on their own.

The best follow-ups are neutral prompts

The safest follow-up questions are short and open. They ask for clarification, not agreement.

Useful prompts sound like this:

• “What happened next?”
• “Can you say more about that?”
• “What made that part difficult?”
• “How did you decide?”
• “What do you mean by ‘slow’?”
• “Can you walk me through that step?”

These are boring on purpose. They keep the participant in charge of the story, which is the point of a qualitative interview [1]. They also reduce the interviewer’s chance of steering the answer with wording, tone, or selective pressure [2][3].

A good neutral probe often targets one of three things: sequence, definition, or example.

Sequence. “What happened after that?” gets the timeline back on the table.

Definition. “When you say ‘confusing,’ what does that look like?” stops everyone from using the same word to mean different things.

Example. “Can you give me a recent case?” moves the conversation from general opinion to a specific incident.

Build follow-ups around the participant’s words

The cleanest follow-up is often a repeat of the participant’s language with one gap left open.

If they say, “I just bounced out of the flow,” ask, “What pulled you out of the flow?”

If they say, “The setup was annoying,” ask, “What part of the setup was annoying?”

If they say, “I usually check somewhere else first,” ask, “Where do you check first?”

This is not trick phrasing. It is a way to avoid translating their experience into your preferred vocabulary. Interview guides are supposed to prompt, not script the respondent’s story [1].

The mistake to avoid is jumping from their words to your theory. If someone says “I was worried about the price,” do not follow with “Was the pricing page unclear?” Those are not the same thing. The first is a feeling. The second is a diagnosis. If you ask the diagnosis too soon, you may get a false precision that sounds useful and is not.

When you need to test a hypothesis, do it after the story

Researchers often want interviews to answer a specific question. That is fine. If you already suspect onboarding is the issue, every probe starts to sound like a search for proof.

A better pattern is to let the participant finish the account first. Then ask the neutral follow-ups that fill in the missing pieces. Only after that should you ask a direct comparison or a hypothesis-shaped question, and even then keep it explicit.

For example:

• “Earlier you mentioned dropping off during setup. What happened at that point?”
• “Was there anything else going on then?”
• “If you had to name the hardest step, what would it be?”

Watch for the hidden leading question

Some follow-ups are leading even when they look polite.

• “Why was that frustrating?” assumes frustration.
• “Did you find that part confusing?” assumes confusion.
• “How badly did that slow you down?” assumes slowdown.
• “Would you say that was the biggest pain point?” assumes priority.

These are common because they feel efficient. They are not. They collapse the range of possible answers before the participant gets to speak.

If you need the same topic covered across interviews, keep the probe neutral and consistent. Ask, “How would you describe that part?” then, if needed, “What made you describe it that way?” That gives you the topic without naming the conclusion [1].

A simple test for every follow-up

Before you ask a probe, check whether it does one of these things:

• asks for a missing fact
• asks for an example
• asks for a definition
• asks for a timeline
• asks for a contrast

If it instead contains your preferred explanation, rewrite it.

“Was the checkout too long?” becomes “What happened during checkout?”

“Did the email feel spammy?” becomes “What about the email stood out?”

“Was the handoff confusing?” becomes “How did the handoff work for you?”

That small rewrite often changes the whole interview. You get fewer yeses and nos. You get more usable detail.

Good interviews do not depend on clever traps. They depend on careful listening and follow-ups that stay out of the way.

Sources

1. Some Strategies for Developing Interview Guides
2. Market research interviews: How to avoid bias & leading questions | MaRS
3. Bias and Debiasing Strategies in Qualitative Data Collection | Philosophy of Science | Cambridge Core