Behavioral interview questions to ask candidates

Ten behavioral questions worth asking

1. Tell me about a time you had to deliver bad news to a stakeholder.
2. Describe a project that did not go the way you expected.
3. Tell me about a time you disagreed with a decision and had to implement it anyway.
4. Describe a time you had to learn something quickly to finish a piece of work.
5. Tell me about a time you received feedback you did not agree with.
6. Describe a situation where you had to say no to someone.
7. Tell me about a time you noticed a problem nobody had asked you to fix.
8. Describe a time you worked with someone difficult. What did you change?
9. Tell me about a deadline you missed.
10. Describe the last time you asked for help.

Ask for an event, not a tendency

The single change that makes behavioral interviewing work is grammatical. “How do you handle conflict?” invites the candidate to describe the person they believe themselves to be. “Tell me about the last time you were in conflict with a colleague” forces them to retrieve something that happened. The first is unfalsifiable; the second has dates, names and consequences you can probe. If a candidate answers an event question with a tendency answer, ask again for a specific occasion — politely, but ask.

The follow-ups that do the real work

Most of the signal in a behavioral interview comes after the first answer. Candidates arrive with three or four polished stories; your follow-ups determine whether you learn anything beyond them.

• “What did you specifically do?” The most important question in behavioral interviewing. Ask it every time the answer is in the plural.
• “What else did you consider?” Reveals whether a decision was reasoned or reflexive.
• “What happened afterwards?” Rehearsed stories often end at the decision. Real ones have consequences, and sometimes the consequences were mixed.
• “What would you do differently?” Separates reflection from narration.
• “How did the other person see it?” Particularly useful on conflict questions. Candidates who cannot represent the other side fairly usually did not resolve it.

Scoring: what a strong answer contains

• A specific situation, with enough detail that it could not be about any other project.
• The candidate's own actions, separable from the team's.
• A decision with a trade-off they can name, rather than a sequence of events.
• An outcome, including when it was not good.
• Something learned that is concrete enough to act on, not “I learned the importance of communication.”

A rehearsed answer is not automatically a bad one

Candidates prepare, and preparation is not deception — it is evidence that they take the conversation seriously. The thing to probe is not polish but depth. A rehearsed story that survives four specific follow-ups is a real story told well. One that dissolves into generalities on the second follow-up was never there. Judge the follow-ups, not the opening.

Behavioral questions by what you need to learn

• Ownership — a time they noticed a problem nobody asked them to fix. Tells you whether they operate inside the task or around it.
• Resilience — a project that did not go as expected, or a deadline they missed. You are listening for what they did when it went wrong, not whether it went wrong.
• Influence without authority — a time they changed someone's mind, or failed to. Particularly important for senior individual contributors, who have to lead without reporting lines.
• Coachability — feedback they did not agree with. Candidates who can represent the feedback fairly before disagreeing with it are usually the ones who act on it.
• Self-awareness — the last time they asked for help. A candidate who cannot remember is either very junior or not reflecting.

Two failure modes to watch for in yourself

The first is the halo effect: one impressive answer early makes every later answer sound better, which is why writing a score per question rather than per candidate matters. The second is similarity bias — rating candidates higher when their stories resemble your own career. Both are mitigated by the same mechanism: decide what a strong answer looks like before the interview, not while listening to one. See how to reduce interview bias.

Score independently, then discuss

Write your assessment before the debrief, not during it. The first strong opinion voiced in a debrief anchors everyone else's, which quietly destroys the independence that made multiple interviewers worth having. See how to run an interview debrief.

Where Interviewer Mode fits

Disclosure: we make CoPilot Interview, which includes Interviewer Mode for the hiring side and a real-time assistant used by candidates. We have an interest in you using the first, and our view of the second is that interviews should be designed to measure something an assistant cannot fake. Weigh the advice below accordingly.

Frequently asked questions

What are good behavioral interview questions to ask candidates?

Questions that ask for a specific past event: a time they delivered bad news to a stakeholder, a project that did not go as expected, a decision they disagreed with but had to implement, a deadline they missed. Avoid “how do you handle X” phrasings, which ask for a self-assessment rather than evidence.

How do I know if a candidate is giving a rehearsed answer?

Follow up with specifics: what did you personally do, what else did you consider, what happened afterwards, how did the other person see it. A real story holds up under four follow-ups and gains detail. A rehearsed one becomes vaguer as you probe. Preparation itself is not a negative signal.

Do behavioral interviews actually work?

Structured behavioral interviews, scored against a rubric, predict job performance considerably better than unstructured conversation. The structure is what does the work — the same questions for every candidate, independent scoring, and evidence recorded before discussion.

How many behavioral questions should I ask?

Three or four in a 45-minute round. Each needs room for several follow-ups, and the follow-ups are where the evidence is. Asking eight questions shallowly produces less signal than asking three properly.