A Practical Guide to Structured Interviews and Scorecards

Ask five interviewers what they thought of a candidate and you will often get five different answers, each delivered with total confidence. One liked the candidate’s energy. Another felt they “weren’t senior enough.” A third spent most of the hour talking about a shared former employer and came away impressed without being able to say why.

Structured interviews exist to fix this. The idea is simple: every candidate for a role is asked the same core questions, and every answer is scored against the same written criteria. Decades of research in industrial psychology show that structured interviews predict job performance far better than unstructured conversations, and they reduce the influence of bias along the way. Yet many teams still treat them as bureaucracy. This guide shows how to make them practical.

What “structured” actually means

A structured interview has four ingredients. Leave any of them out and you are back to gut feel with extra paperwork.

1. Defined competencies. A short list of the skills and behaviours the role requires.
2. Standard questions. The same core questions for every candidate, mapped to those competencies.
   • Follow-up probes are allowed, but they should also be written down in advance.
   • Interviewers should not swap questions between candidates.
3. A scoring rubric. Written descriptions of what a weak, acceptable and strong answer looks like.
4. Independent scoring. Each interviewer submits their scores before hearing anyone else’s opinion.

The fourth point matters more than most teams realise. Once a senior person says “strong hire” in a debrief, everyone else’s memory of the interview starts to shift.

Choosing competencies

Start from the job, not from a generic list. Sit down with the hiring manager and ask: what will this person need to do well in the first year, and what has caused people in similar roles to struggle? You should end up with four to six competencies. More than that and interviewers can’t keep them in their heads.

For a recruiting coordinator role paying £34,000–£40,000 in London or $58,000–$68,000 in the US, like the one we advertised in our Recruiting Coordinator listing, the list might look like this:

• Organisation and attention to detail
• Written communication
• Candidate empathy
• Working with ambiguity
• Tooling and process improvement

Each competency should be owned by one interview stage, with at most one backup. If three interviewers are all assessing “communication”, you are spending three hours on one skill and none on others.

Writing good questions

The best structured interview questions come in two forms.

Behavioural questions

These ask about something the candidate has actually done: “Tell me about a time you had to reschedule a full interview loop at short notice. What happened, and what did you do?” Past behaviour in similar situations is one of the better predictors of future behaviour, and specific stories are harder to bluff.

Situational questions

These describe a realistic scenario and ask how the candidate would handle it: “A candidate in Sydney emails at 6:40 a.m. your time to say the video link isn’t working, and their interview started 10 minutes ago. Walk me through what you do.” Situational questions work well for people with less experience, who may not have a past example to draw on.

Avoid brainteasers, hypothetical “what animal would you be” questions and anything that tests trivia the person could look up in 30 seconds. They feel rigorous but predict very little.

“When we switched to structured loops, the biggest change wasn’t the questions. It was that interviewers had to write down why they gave a 2 instead of a 3. That one habit cut our post-debrief disagreements in half.”

Mireille Dubois, Director of People Operations at Kestrel Health

Building the scorecard

A scorecard turns your competencies and rubric into a form interviewers fill in right after each session. Keep the scale short. A four-point scale works well because it forces a decision; there is no comfortable middle option to hide in.

Score	Label	What it means	Typical share of candidates
1	Strong no	Clear evidence the candidate can’t meet the bar	15–20%
2	Leaning no	Some evidence, but significant gaps or concerns	30–35%
3	Leaning yes	Meets the bar with minor gaps	30–35%
4	Strong yes	Exceeds the bar with specific, convincing evidence	10–15%

If almost every candidate is scored a 3, your rubric is too vague. If almost nobody gets a 4, either your bar is unrealistic or your interviewers are being cautious. Reviewing the distribution once a quarter, for example at the end of Q3 2026, is a quick way to spot calibration problems.

Many teams store scorecards in their applicant tracking system, but it helps to keep the template itself in a plain file so changes are versioned and reviewed like any other process document.

role: Recruiting Coordinator
version: 2026-07-01
stage: Logistics & Communication (45 min)
interviewer_instructions: >
  Ask Q1 and Q2 to every candidate. Use only the listed probes.
  Submit scores within 2 hours, before reading other feedback.
competencies:
  - id: organisation
    question: "Tell me about a time you rescheduled a full interview loop at short notice."
    probes:
      - "What did you communicate to the candidate, and when?"
      - "What would you do differently?"
    rubric:
      1: "No concrete example, or blames others for the outcome."
      2: "Example given, but actions are vague or reactive."
      3: "Clear actions, candidate kept informed, outcome described."
      4: "Clear actions plus a process change that prevented repeats."
  - id: written_communication
    question: "Rewrite this rejection email so it is kind, clear and brief."
    rubric:
      1: "Unclear or unkind tone."
      4: "Under 120 words, specific, warm and final."
recommendation: [strong_no, leaning_no, leaning_yes, strong_yes]

Common scoring traps

Even with a rubric, a few predictable biases creep into scores. Knowing their names makes them easier to catch.

• Halo effect: one impressive answer early in the interview lifts every score that follows.
• Similarity bias: candidates who share an interviewer’s background, school or hobbies get the benefit of the doubt.
• Central tendency: cautious interviewers score everyone a 2 or 3 to avoid being wrong.
• Recency: the last ten minutes of the conversation outweigh the first forty.

The best defence is to take notes during the interview and score each competency separately, immediately afterwards, by reading the notes back rather than relying on a general impression. Some teams also ask interviewers to quote one sentence from the candidate as evidence for every score of 1 or 4.

Training your interviewers

A scorecard is only as good as the people using it. Before anyone joins a structured loop, they should complete a short training session. In our experience, 90 minutes is enough to cover the essentials, followed by two shadowed interviews.

Training should cover:

• Why structure matters, with a few real examples from your own past hires.
• How to ask the questions without leading the candidate toward the answer.
• Note-taking: writing down what the candidate said, not your interpretation of it.
• Scoring: practising on two or three recorded or written sample answers, then comparing scores as a group.

The calibration exercise in the last step is the most valuable part. When five interviewers score the same answer as a 1, a 2, a 3, a 3 and a 4, the discussion that follows teaches more than any slide deck.

Running the debrief

The debrief is where structured hiring most often falls apart. Here is a format that works for loops of three to six interviewers:

1. Scores are submitted first. Nobody joins the meeting until their scorecard is in.
2. The facilitator reads scores aloud, starting with the most junior interviewer.
3. Discuss disagreements only. If everyone scored a competency 3 or 4, move on.
   • Ask the outliers to share their evidence, not their impression.
   • Allow people to change their score, but record the original.
4. The hiring manager decides, using the scores as input rather than a vote count.
5. Write up the decision in two or three sentences, linked to the competencies.

Keep debriefs to 30 minutes. If you regularly need longer, your scorecards probably aren’t capturing enough evidence.

Measuring whether it works

Structure is an investment, so check that it pays off. Six months after a hire’s start date, ask their manager to rate performance on the same competencies used in the interview. Over time, you can see which stages predict success and which don’t. One mid-sized company we work with found that its take-home exercise had almost no correlation with six-month performance, so it dropped the stage and shortened its process from 21 days to 14.

Track a few simple numbers each quarter:

• Time from application to offer, in days
• Offer acceptance rate, as a percentage
• Score distribution per interviewer
• Six-month performance ratings for new hires

Getting started this month

You don’t need to restructure every role at once. Pick one high-volume position, write four or five competencies, draft questions with the hiring manager and run the new loop for the next 10 candidates. Then review.

If you’re hiring now, you can post a job on HireWise and link your interview process directly in the listing. Candidates notice. For more on what experienced applicants expect to see before they apply, read our guide to writing job descriptions for senior engineers, or browse open recruiting roles to see how other teams describe their loops. Questions about managing applicants on the platform are covered in the help center.