Probing questions for interviewers: 40 follow-ups organized by purpose

Probing questions are the follow-ups that turn a candidate's first answer into evidence you can score. Almost every useful probe does one of six jobs: clarify a term, get a specific example, separate what the candidate did from what the team did, go deeper into their reasoning, get the result, or test a claim. Below are 40 follow-ups organized by those purposes, a table for choosing the right one from what you just heard, a worked example, and the probes to avoid. They work with any question in your kit.

Our competency pages, such as interview questions for active listening, give probes tied to one question each. This page is the general library you reach for when the answer goes somewhere the kit did not predict.

Why do probing questions decide the score?

Most first answers are summaries: a habit, a team achievement, a conclusion without the reasoning. The score comes from the follow-ups. The U.S. Office of Personnel Management's guidance on structured interviews describes the most structured interviews as asking every applicant the same pre-defined lead and probe questions, which is why the probes below are worth writing into the kit rather than improvising.

Probing also needs to be even-handed. In a field study of 79 audiotaped interviews, Dougherty, Turban and Callender (Journal of Applied Psychology, 1994) found that interviewers followed up positive first impressions by showing positive regard, selling the company, giving job information and gathering less information. Probe the candidate you like as hard as the one you have doubts about.

Which probe should you use?

Clarify: what exactly do they mean?

• "When you say you owned it, what did that involve day to day?"
• "Who was 'they' in that story?"
• "How big was it: the team, the budget, the number of clients?"
• "What was your role at the time, and who did you report to?"
• "Can you explain that the way you would to someone outside your field?"
• "What does that term mean in your company?"

Use when a word could mean several things. Clarifying is not doubting; ask it in a neutral tone.

Get a specific example: did it actually happen?

• "Can you think of one specific time? When was it?"
• "When was the last time that happened?"
• "Pick the hardest one. What was going on?"
• "Walk me through that day, from when you first heard about it."
• "It doesn't have to be a big project. A small example is fine."
• "Has anything like that actually happened to you?" (after a hypothetical answer)

Use when the answer describes a habit, a policy or what they would do.

Separate "I" from "we": what did the candidate do?

• "What was your part, specifically?"
• "What did you do that nobody else on the team did?"
• "Who made the final decision?"
• "What did you say to them, roughly in your own words?"
• "If I asked your manager what you contributed, what would they say?"
• "What would have happened if you had not been involved?"
• "Which part would you have done differently if it had been entirely up to you?"

Use when the story is told in the plural. Some candidates say "we" out of modesty or because their culture discourages self-promotion, so ask; do not mark them down for the pronoun.

Go deeper: how did they think about it?

• "Why that approach rather than another?"
• "What other options did you consider?"
• "What did you know at the time, and what did you only find out later?"
• "What was the hardest part?"
• "What did your choice give up?"
• "Who disagreed, and what was their argument?"
• "What would have made you change course?"

Use when you have the actions but not the judgment behind them.

Get the result: what happened in the end?

• "What happened in the end?"
• "How did you know it worked?"
• "What changed as a result, and by roughly how much?"
• "Is it still in place?"
• "What would you do differently next time?"
• "What did you learn that you have used since?"

Use when the story stops at the action. A result that did not go well is still evidence; ask what they did next.

Test a claim: how do they know?

• "How did you measure that?"
• "What was it before you started?"
• "Walk me through how you got to that number."
• "What went wrong along the way?"
• "Who else was involved who could describe it?"
• "What did the people affected think of it?"

Use when a claim matters to the score: a revenue figure, a percentage improvement, "the best in the region". Test claims that bear on the decision, not every number, and ask the same depth of every candidate.

Close the topic: how do you move on politely?

• "Is there anything about that you'd like to add before we move on?"
• "That's helpful, thank you. I want to make sure we get to a few other areas."

The second is the polite way to stop a long answer; candidates are usually relieved to be steered.

How many probes should you ask?

Two or three per main question is typical. Stop as soon as you can place the answer on your anchors: you know whether each element of a strong answer is present or absent. Five or more follow-ups on one answer turns into a cross-examination and spends the time another competency needed. Decide the depth in advance and keep it the same for every candidate; probing only the candidates you doubt is a bias in itself.

How do you probe a nervous, brief or rambling candidate?

• Nervous: give them time. Silence for a few seconds after a question is normal, and the first probe can be an easy one: "Take your time. Where were you working then?"
• Very brief: short answers are not necessarily weak ones. Ask for one detail at a time ("What happened next?") rather than "Can you say more?", which gives them nothing to hold on to.
• Rambling: interrupt kindly and narrow the question: "Let me stop you there. What was your part in that?" A long answer that never reaches the candidate's own actions is still a 2.
• Speaking in a second language: rephrase in plainer words rather than repeating the same sentence, and score the content of the answer, not the fluency, unless the language is a requirement of the job.

Which probes should you avoid?

• Leading probes: "So you'd say you stayed calm?" invites a yes. Ask "How did you respond?" instead. There are 30 more rewrites in leading questions in interviews.
• Double probes: "Why did you do that and how did it go?" Ask one at a time.
• Accusing probes: "Why didn't you just escalate?" Try "What options did you have at that point?"
• Personal probes: following a mention of children, health, religion or age. Acknowledge it, do not follow up; see illegal interview questions.
• Probes outside the competency: interesting detours that will not change the score.

A worked example

How do you use probes in a live interview?

1. Under each question in your kit, write the two or three probes you expect to need.
2. During the answer, note which elements of the anchors you have heard and which are missing.
3. Choose the probe for the biggest gap, using the table above.
4. Write the candidate's answer in their own words, so the score can be traced to it; see evidence-based interview feedback.

If you use Interview Signal, it suggests follow-ups during the call, each labelled with its reason (probe, clarify or evidence); how that works is in how follow-up suggestions work. The choice of which one to ask stays with you.