Writing Structured Interview Questions

A structured interview is only as good as its questions. Writing eight good ones takes an afternoon and they are reusable indefinitely.

Once the criteria described in “Writing Structured Interview Questions” are fixed, the practical challenge is running them consistently across assessors. A coordinator can use the time-recording overview to plan review time, compare the effort spent at each stage and spot avoidable delays, without confusing administrative efficiency with the quality of the evidence collected from candidates.

For an independent benchmark, compare the process with CIPD selection methods guidance; the useful test is whether the local method remains job-related, proportionate and explainable.

Start from the critical incidents

The job analysis should have produced two or three stories of somebody handling a situation notably well or badly.

Each of those becomes a question.

This is the shortcut: questions drawn from real situations are automatically relevant, and generic competency questions are not.

The behavioural form

"Tell me about a time when..." followed by a specific situation type.

Not "are you good at X" and not "how would you approach X", which invite the answer the candidate thinks you want.

Follow up for specifics: what exactly did you do, what happened, what would you do differently.

The follow-ups are where the evidence is, which is why they should be written down rather than improvised.

The situational form

"Here is a situation. What would you do, and why."

Useful where candidates lack directly relevant history, which makes it fairer for career changers and early-career applicants.

Score the reasoning rather than the answer, because several approaches are usually defensible and the thinking is what distinguishes.

Questions that do not work

Hypotheticals about extreme situations candidates will never face.

Anything answerable from the job advertisement.

"What is your greatest weakness", which measures preparation for interviews.

Brain teasers, which have no demonstrated relationship with performance.

And anything about personal circumstances, which is irrelevant and in many jurisdictions unlawful.

How many

Five to eight for a substantial interview.

Each mapped to exactly one requirement.

More than eight and the interview runs long, assessors tire, and the later questions get scored carelessly.

Writing the scoring guide

For each question: what a strong answer contains, what an adequate one contains, what is missing from a weak one.

Written in terms of evidence rather than impression: "describes a specific situation with their own actions and a measurable outcome" rather than "seems capable".

Test it by scoring three hypothetical answers yourself. If you cannot apply your own rubric, neither can anybody else.

Reusing them

The same set works for every candidate for that role and for future rounds.

Review annually, and after any substantial change to the role.

And rotate specifics occasionally if questions are circulating publicly, which happens and matters less than people think — a rehearsed answer to a behavioural question still describes something the candidate did.

What to check

Did your questions come from real situations in this role?

Is each one mapped to a requirement?

Are the follow-up prompts written down?

And could two people score the same answer the same way using your guide?

The point

Draw interview questions from real critical incidents rather than from a competency list.

Questions from real situations are automatically relevant.

Underlying all of this

Almost everything in this collection reduces to one discipline: write down what the job requires, assess that thing directly, record the evidence, and look at your own outcomes afterwards. None of it requires buying anything, and organisations that do those four things consistently outperform ones running longer processes built from instruments chosen before the requirements were known.

The recurring pattern

The recurring failure across every section here is the same: measuring what is convenient rather than what matters, then never checking whether it predicted anything. The check is an afternoon of work once a year, and it is the step that separates a process that improves from one that merely persists.