Probing Questions for User Interviews: 40+ Examples to Get Deeper Answers

What probing questions are (and why they make or break an interview)

Bottom line up front: a probing question is a follow-up that digs beneath a participant's first answer to reach the reasoning, emotion, or specifics underneath it. The first answer in an interview is almost always the surface — "it was fine," "I just liked it." The insight lives one or two probes deeper. Mastering probes is the single highest-leverage interviewing skill, and it is the one Koji's AI interviewer automates so every conversation goes deep, not just the ones run by your best researcher.

The six types of probing questions

1. Elaboration probes

Get the participant to say more.

• "Can you tell me more about that?"
• "What do you mean by 'frustrating'?"
• "Walk me through that step by step."
• "What else was going on at that moment?"

2. Clarification probes

Resolve ambiguity before you misinterpret.

• "When you say 'it,' what specifically are you referring to?"
• "Help me understand — was that before or after you signed up?"
• "Can you give me an example of that?"
• "How are you defining 'expensive' here?"

3. Specificity / example probes

Move from the abstract to the concrete.

• "Tell me about the last time that happened."
• "Can you remember a specific instance?"
• "What did you do next?"
• "Show me how you would normally do that."

4. Reasoning / "why" probes

Uncover motivation and cause.

• "Why was that important to you?"
• "What made you choose that option over the others?"
• "What were you hoping would happen?"
• "What would have happened if you hadn't done that?"

5. Emotion probes

Surface how the experience felt.

• "How did that make you feel?"
• "What was the most frustrating part?"
• "What surprised you about it?"
• "When were you most confident — or most worried?"

6. Contrast / anchor probes

Pin down magnitude and trade-offs (especially for ratings).

• "You rated it a 6 — what would have made it a 9?"
• "How does this compare to how you did it before?"
• "What almost made you give up?"
• "If you could change one thing, what would it be?"

The simplest, most powerful probe of all is silence — pausing for two or three seconds after an answer. People fill the gap with their real thoughts. (Human-only superpower; AI substitutes a gentle "anything else?")

A worked example: from surface to insight

Q: How was onboarding? A: Yeah, it was okay. Probe (elaboration): What does "okay" mean here — what stood out? A: I got stuck connecting my data source. Probe (specificity): Walk me through exactly where you got stuck. A: The auth popup closed and I didn't know if it worked. Probe (emotion): How did that feel in the moment? A: Honestly I almost gave up — I thought I'd broken it.

Three probes turned "it was okay" into a specific, fixable, emotionally weighted insight. That is the entire game.

Common probing mistakes

• Leading the witness. "So that was confusing, right?" plants the answer. Stay neutral: "What was that like?" (See avoiding bias in interviews.)
• Stacking questions. Asking three probes at once lets the participant pick the easy one. Ask one, then wait.
• Accepting the first answer. "Convenience" is a label, not an insight. Probe it.
• Probing everything. Not every answer needs three follow-ups; over-probing exhausts people. Probe what matters to your research goal.
• Fear of silence. Filling every pause robs you of the participant's best, most considered thoughts.

How Koji probes automatically — and consistently

The hard truth about manual interviewing is that probing quality is wildly uneven. Your senior researcher gets gold; a rushed teammate accepts "it was fine" and moves on. Koji's AI interviewer applies disciplined probing to every conversation, at any scale.

Each question in a Koji study carries its own probing configuration:

• maxFollowUps — how deep to probe (0 = just ask; 1–3 = probe depth). Set it per question so high-value topics get deeper digging.
• instructions — specific guidance for what the AI should probe toward (e.g., "probe for the underlying job, not the feature").
• anchor — for scale questions, the AI automatically asks the contrast/anchor probe: "You said a 6 — what would make it a 9?"

Because this is configured at the question level, you get the depth of a skilled interviewer across hundreds of conversations simultaneously — voice or text — without a moderator burning out on interview number forty.

Probing works best on structured questions

Probing depth pairs directly with Koji's structured questions — six first-class types that decide what gets asked and how it is probed:

• open_ended — the natural home of probing; free-form answers get AI follow-ups coded into themes
• scale — pair with the anchor probe to learn what would move the number
• single_choice / multiple_choice — probe why they chose what they chose
• ranking — probe the reasoning behind the top and bottom items
• yes_no — always worth a "why?" probe

See the structured questions guide to set probing depth per question. The probed answers are then coded into themes automatically, so the depth you capture actually shows up in the report — not lost in a transcript no one re-reads.

Probing in voice vs. text interviews

Good probing adapts to the medium. In voice interviews, probes feel conversational and silence is a real tool — a two-second pause invites elaboration. In text interviews, you lose silence but gain something else: participants can take their time and often write more considered answers, and a well-timed typed follow-up ("Interesting — can you give a specific example?") rarely feels intrusive. Koji runs disciplined probing in both modes through the same per-question configuration, so a study mixing voice and text stays consistent in depth. Many teams run voice for emotionally rich topics and text for reach, knowing both will be probed and coded the same way.

Match probe depth to your research goal

Not every question deserves three follow-ups. Tune probing to what the study needs:

• Discovery and "why" research — set deeper probing (maxFollowUps 2–3) on your core open-ended questions; this is where laddering and the 5 Whys live.
• Satisfaction and rating studies — enable the anchor probe on scale questions so every score comes with the reason it is not higher.
• Quick validation or screening — keep probing light (0–1) so the conversation stays short and completion stays high.

The skill is knowing where the insight is and spending your probes there. Encode that judgment once in a Koji study and the AI applies it identically to every participant — the consistency a tired human moderator on interview number forty simply cannot match.

Quick-start: a reusable probing ladder

For almost any answer, this sequence works:

1. Elaborate — "Tell me more about that."
2. Get specific — "When did that last happen?"
3. Find the why — "Why did that matter?"
4. Capture emotion — "How did that feel?"

Use it as your manual cheat sheet — or encode it once in a Koji study and let the AI run it on every participant.

Related Resources

• Structured Questions in AI Interviews — set probing depth per question
• The Laddering Technique — probing toward deeper motivations
• The 5 Whys Technique — root-cause probing
• The Mom Test: Asking Questions That Get Honest Answers — neutral, bias-free questioning
• How AI Interviewers Work — the engine behind automatic probing
• Can You Trust AI Interviewers? — avoiding leading and bias