Hiring · July 20, 2026 · 9 min read

Structured Interviews: A Guide to Fairer Hiring

Structured interviews are among the strongest predictors in hiring. This guide covers what makes an interview structured and how to build, score and run one.

← Part of The five pillars of hiring: what assessments measure

For any hiring manager or talent leader, the interview is usually the most trusted step in the process, and, left unstructured, one of the least reliable. The stakes are real: an interview that swings on rapport and gut feel quietly rewards candidates who are confident, articulate and similar to the interviewer, predicts actual job performance weakly, and leaves a decision you cannot defend if it is challenged. Structured interviews fix this. By asking every candidate the same job-relevant questions and scoring against the same anchored rubric, you turn the interview from a conversation into a measurement, and one of the strongest predictors available in hiring.

This guide covers what "structured" actually means, how to build and score an interview that holds up, and how to sit it on top of real evidence so the whole evaluation gets sharper. The aim is not more bureaucracy; it is more signal per hour of everyone's time.

What 'structured' actually means

A structured interview has two properties, and only two. Every candidate gets the same job-relevant questions, and every answer is scored against the same anchored rubric. That is it. Those two constraints remove most of the noise, the tangents, the halo effects, the gut calls, that make unstructured interviews inconsistent from one candidate to the next. Everything else people associate with "structure", panels, competency frameworks, note-taking, is useful scaffolding, but it is not what makes the interview work. Consistency of questions and consistency of scoring do.

Why unstructured interviews under-perform

An unstructured interview measures the interviewer as much as the candidate. Free conversation invites first impressions to harden within minutes, and everything after that is read as confirmation. It rewards fluency and similarity over competence, and because each candidate is asked different things, there is no common yardstick to compare them against. The result is a step that feels insightful and predicts weakly. Structure does not remove human judgement; it channels it, by fixing the questions and the scale so judgement is applied to the same evidence for everyone.

The word "feels" is doing a lot of work there. Interviewers reliably rate their own accuracy far higher than it is, because a confident, articulate candidate produces a confident, articulate impression, and that impression is mistaken for insight. This is exactly why the interview is so trusted and so misleading at once: the experience of interviewing is persuasive even when the outcome is close to noise. Structure is the antidote precisely because it does not depend on the interviewer feeling accurate; it makes the questions and the scoring do the work that memory and instinct do badly.

The most common failure modes

• The halo effect: one strong answer, or a warm first minute, colours every later judgement so the whole interview reads as a nine.
• Similarity bias: rating candidates higher when they share a background, school or communication style with the interviewer.
• Anchoring in the debrief: the first opinion voiced sets the group's baseline, and everyone adjusts toward it rather than reporting what they actually observed.
• Recency and fatigue: the last candidate of the day, or the one after a bad one, is scored against a shifting bar rather than a fixed one.

How to build one

Building a structured interview is mostly work you do before anyone walks in the room. The four steps below are the difference between a scorecard that shapes the conversation and one that gets filled in from memory afterwards.

• Define the competencies. Decide what the role actually requires before you write a single question. The five pillars, cognitive, domain, situational, behavioural and AI fluency, are a useful frame, and the skills-based hiring guide covers mapping them to a specific role.
• Write job-relevant questions. Favour realistic scenarios over trivia or brain-teasers, and ask about work the person will really do. Situational judgement and behavioural questions each probe a different pillar.
• Anchor the scoring. Agree what a strong, average and weak answer looks like before the interview, with short example behaviours at each level, so scores mean the same thing across interviewers.
• Score independently. Have interviewers rate before they discuss, to avoid the loudest voice setting the consensus and the group anchoring on the first opinion voiced.

Getting the questions right

The most common mistake is writing questions that sound rigorous but test the wrong thing, riddles, gotchas, or hypotheticals so abstract that they reward glibness. A good structured question describes a situation the candidate will plausibly face and asks how they would handle it, or asks them to walk through something they have actually done. Keep the set small enough to score carefully. Ten shallow questions produce worse signal than four probed in depth.

Two question types carry most structured interviews. Behavioural questions ask what a candidate actually did in a past situation, on the reasoning that past behaviour predicts future behaviour better than stated intentions. Situational questions pose a realistic scenario and ask how they would approach it, which is useful when the candidate has not yet held the role. Both should tie back to a specific competency you defined, and both need agreed follow-up probes so an interviewer can dig past a rehearsed opening answer without going off-script. If a question does not map to a competency, cut it; it is filler that dilutes the score.

Anchoring the scale

An unanchored five-point scale is barely better than gut feel, because a four means something different to every rater. Anchored rating scales fix this by describing, in advance, what a strong, average and weak answer looks like for each question, ideally with a short example behaviour at each level. The anchors do two jobs: they make scores comparable across interviewers, and they force you to articulate what "good" means before the room's chemistry can define it for you. Writing anchors is tedious, and it is also where most of the fairness and predictive power comes from.

A worked example

Take a question probing how a candidate handles conflicting priorities: "Tell me about a time two stakeholders wanted incompatible things from you at once. What did you do?" A weak answer describes the conflict but not the candidate's own actions, or blames others. An average answer explains what they did but not why, or resolves the conflict by simply deferring to seniority. A strong answer surfaces the trade-off explicitly, describes how they weighed it, names who they consulted, and reflects on what they would do differently. Written down like that, three interviewers scoring the same answer will land close together, which is the whole point.

Notice what the anchors are not doing: they are not scripting the interview or forcing a single "right" answer. Two candidates can score highly with completely different approaches, as long as each shows the reasoning the competency calls for. That is the difference between a rubric and a checklist. A rubric rewards good judgement wherever it appears; a checklist rewards saying the expected keywords. Structure, done well, expands the range of good answers it can recognise rather than narrowing it.

Interviews work best on top of evidence

A structured interview gets sharper when it is not starting cold. Run a work sample first, and the interview becomes a deep dive into what the candidate actually produced: their reasoning, their trade-offs, the parts they would change, rather than a fresh guess based on how they describe themselves. The assessment surfaces the evidence; the interview interprets it. This split also plays to each step's strength: a job-relevant assessment is the fairest way to measure whether someone can do the work, and a structured interview is the best way to understand how and why they did it.

This division of labour also makes each step honest about what it is good at. Ask a candidate in an interview whether they can debug a tricky concurrency issue and you get a story; hand them a realistic task and you get an answer. Conversely, a work sample cannot easily tell you how someone reasons about ambiguity or explains a trade-off to a non-technical stakeholder, which a structured interview probes well. Running the two in sequence, evidence first, interpretation second, means each stage is asked only to do the thing it does reliably, and the overall decision rests on more than any single impression.

Common objections, briefly

Two objections come up whenever teams move to structure. The first is that it feels robotic and will put candidates off. In practice a well-run structured interview is still a conversation; the interviewer just asks the same core questions and probes freely within them. The second is that it takes too long to set up. It does take upfront work to define competencies and write anchors, but that work is reused across every candidate for the role and every future opening like it, so the cost is paid once and the benefit compounds. Neither objection survives contact with a process that has actually been run this way for a few cycles.

Keep it fair, and keep checking

Structure reduces bias substantially, but it does not eliminate it. A poorly worded question can still favour one group, and a rater can still drift over a long day of interviews. Pair the interview and the assessment with ongoing monitoring: watch outcomes across groups for adverse impact, and treat any imbalance as a prompt to inspect the question or the rater, not to lower the bar. Standardising the process is the first move to reduce bias; measuring the outcomes is how you know it worked. Done together, the two protect fairness across the whole funnel and lift your quality of hire over time.

There is a compliance dimension too. As hiring regulation tightens, the ability to show that every candidate was asked the same job-relevant questions and scored against the same documented standard is increasingly what separates a defensible decision from an indefensible one. A structured process produces that record almost as a by-product: the questions, the anchors and the independent scores are the audit trail. An unstructured interview, by contrast, leaves you reconstructing a decision from memory, which is exactly the position you do not want to be in if a rejected candidate ever asks why.

The candidate's side of it

Structure is not just fairer to the business; it is fairer to the candidate, and it usually reads that way. A clear, job-relevant interview where everyone is asked the same things feels less arbitrary than a rambling chat whose outcome hinges on chemistry. That consistency is a quiet contributor to a good candidate experience, which in turn protects your offer-acceptance rate. Fairness and appeal are not in tension here; the same discipline serves both.

Putting it together

A structured interview is not a heavier interview, it is a more honest one. Define the competencies, write job-relevant questions, anchor the scale, score independently, and sit the whole thing on top of real evidence from a work sample. Then keep checking the outcomes. Each of those steps removes a bit of the noise that makes ordinary interviews unreliable, and together they turn your most trusted hiring step into one you can actually trust.

An unstructured interview measures the interviewer as much as the candidate. Structure moves the measurement back where it belongs.