Write questions from the role's behavioral profile, ask every candidate the same ones, and rate each answer against a defined scale before you discuss the candidate. Structure, not clever questions, is what makes interviews predictive: structured interviews average a validity of .42 compared with .19 for unstructured ones.
Most of us believe we can read people in an interview. The research suggests we are better at feeling confident than at being accurate. The striking finding is not that interviews fail. It is that the same hour, spent differently, more than doubles what an interview can tell you. The difference is not cleverer questions. It is discipline.
Why does interview structure matter so much?
In Sackett et al.'s 2022 re-analysis, structured interviews were the strongest predictor of job performance of any selection method studied, with an average validity of .42. Unstructured interviews averaged .19. Same activity, same time cost, more than double the predictive value. The difference is structure.
The research also shows a wide range around that average. A well-built structured interview can do much better than .42. A poorly built one can do much worse. What makes the difference is how closely the questions track what the job actually needs.
How do you build the questions?
- Start from the role profile. Each question should test one behavior from the profile. If a question doesn't map to a behavior, cut it.
- Ask about real past situations. "Tell me about a time when…" invites evidence. "What would you do if…" invites a rehearsed answer.
- Probe the gaps. Where a candidate's assessment results differ from the role profile, write a question that explores that difference directly.
- Agree on what good looks like. Before interviews start, write down what a strong, acceptable and weak answer sounds like for each question.
Example questions by behavior
| Behavior in the profile | Question |
|---|---|
| Works through ambiguity independently | Tell me about a project where the goal was unclear. What did you do in the first week? |
| Prefers closure and precision | Describe a time you found an error after work had been signed off. What happened next? |
| Builds collaboration across teams | Tell me about a time you needed something from a team that didn't report to you. |
| Stays motivated through routine work | What part of your last role was most repetitive, and how did you keep the quality up? |
How should interviewers rate answers?
Each interviewer rates each answer independently, against the agreed scale, before any group discussion. In the GoodJob process, interviewers rank answers to surface both gaps and natural talents, so the debrief starts from evidence rather than from whoever speaks first. Research summarized by McDaniel and colleagues also suggests that individual structured interviews tend to outperform panel discussions where one strong voice can anchor the room.
Common mistakes
- Letting each interviewer write their own questions.
- Discussing the candidate before ratings are recorded.
- Asking about hypotheticals instead of past behavior.
- Treating rapport as evidence.
Frequently asked questions
What is the STAR method?
STAR (situation, task, action, result) is a way for candidates to structure answers about past behavior. Interviewers can use it as a probe checklist to make sure an answer contains real evidence.
How many questions should a behavioral interview have?
Enough to cover each behavior in the role profile once or twice. For most roles that is a focused set that fits in under an hour.
Should every candidate get the same questions?
Yes for the core questions, so answers can be compared fairly. Follow-up probes can be tailored to each candidate's assessment results.
Related
What is behavioral hiring, and why does it predict performance better?
Behavioral hiring selects people on the work behaviors a role actually needs. Here is how it works, what the evidence says, and how it differs from résumé and gut-feel hiring.
Behavioral assessments vs. personality tests: what's the difference for hiring?
MBTI, DISC and Kolbe describe people. Behavioral assessments predict fit for a specific role. Here is the difference and why it matters for hiring decisions.