The short answer
An interview scorecard template gives every interviewer the same form, so candidates are judged on the job and not on who happened to interview them. A good one rates four to six job-related competencies on a fixed scale, with a written description of what each score looks like. Interviewers write down what the candidate actually said, score each competency on its own, and submit before any group discussion. The U.S. Office of Personnel Management (OPM) says structured interviews built this way have shown a high degree of reliability, validity and legal defensibility, while unstructured interviews tend to have low reliability.
Free interview scorecard template
Copy this into a doc, your ATS or a form tool. Replace everything in [square brackets]. The example competencies suit a customer-facing role, so swap in the ones your job analysis points to.
The 1 to 4 interview rubric
OPM's guide recommends at least three proficiency levels, and suggests aiming for five to seven, with at least three of them labeled. Many hiring teams settle on four because it has no middle option. An interviewer has to decide whether the evidence met the bar or not, which counters what OPM calls central tendency, the habit of rating everyone in the middle. If you prefer five points, the method below works the same way.
Write anchors for each competency, not just the general scale. "Meets" on customer communication and "meets" on SQL describe very different answers, and the specific anchor is what lets two interviewers agree.
How to weight competencies
Not every skill matters equally. Weighting forces the hiring team to agree on that before interviews start, instead of arguing about it in the debrief. A simple method:
- Start from the job. OPM recommends a job analysis first: identify the tasks, the competencies needed to do them, and which of those are needed on day one. A current job description is a good starting point.
- Pick four to six competencies. OPM notes that structured interviews typically assess between four and six, unless the job is unique or very senior.
- Split 100 points. Give the most points to the skills that most separate strong from weak performers in this job, not the ones easiest to talk about.
- Set any must-have floors. If a 1 on a competency means "no hire" whatever the total, write that down now.
Here's a worked example for an account executive, with one candidate's ratings:
Use the total to compare candidates for the same role, never across roles. And treat it as a summary of evidence, not a verdict. Two candidates at 3.1 can have very different profiles, which is exactly what the debrief should discuss.
Scorecard, rubric or evaluation form: what's the difference?
People use these terms loosely, and the overlap causes confusion when a team sets up its process:
- Interview rubric: the scoring key. It describes what each rating looks like for each competency.
- Interview scorecard: the form an interviewer fills in for one candidate, with competencies, questions, ratings, evidence and a recommendation. A good one has the rubric printed on it.
- Interview evaluation form: often the same thing as a scorecard. Some teams use the phrase for a shorter, end-of-process summary that the hiring manager signs.
Whatever you call it, the test is the same: could a different interviewer read the notes and understand why the candidate got each score?
How to fill in a scorecard well
The form only helps if interviewers use it during and right after the conversation. OPM's guidance is specific here:
- Take notes as you go. Record what the candidate said and did, in enough detail to explain each rating later. Short quotes are better than impressions like "seemed sharp."
- Use the same probes for everyone. OPM says follow-up probes should be very similar across candidates, so each person gets the same chance to show what they know.
- Rate immediately. Score each competency right after the interview, while the notes still make sense, and tie every rating to an example.
- Rate independently, then discuss. On a panel, OPM's guidance is that each member rates alone first, then the panel explores the reasons for any differences and reaches a consensus.
- Keep the records. OPM lists interview questions, scores and notes for each candidate among the documents to retain, along with the guides and rating scales used.
Calibration tips for hiring teams
- Score one recorded interview together. Before a new role opens, have every interviewer score the same mock or recorded interview, then compare. Disagreements of two points or more mean the anchors need rewriting.
- Check each interviewer's spread. After ten or so interviews, look at each person's average. Someone who gives almost every candidate a 4 (leniency) or a 1 (strictness) needs a conversation, not a new form.
- Watch for one score driving the rest. OPM calls this the halo effect: a strong rating on one competency pulling up the others. Scoring each competency from its own evidence is the fix.
- Review hires after six months. Compare scorecard ratings with how new hires are doing. If high scorers on one competency aren't performing better, that question or anchor isn't measuring what matters.
- Use the same interviewers for a role where you can. OPM suggests keeping the same interviewers across all candidates when feasible, to keep ratings consistent.
How scorecards reduce bias
A scorecard doesn't remove bias by itself, but it closes the gaps where bias usually gets in. OPM's guide lists common rating errors, including rater bias (letting views about a group affect a rating), similar-to-me (rating people higher because they resemble you), first impressions, and contrast effects (judging a candidate against the one before rather than against the standard). Structure counters each one: the same questions keep the conversation on the job, written anchors replace gut feel, evidence notes make every rating checkable, and independent scoring stops the loudest voice from setting the result.
Keep the form itself clean, too. The EEOC says pre-employment questions should be limited to what's essential for deciding whether a person is qualified, that questions about race, sex, national origin, age and religion are irrelevant to that decision, and that employers may not ask about disability before making an offer. Nothing on your scorecard should invite notes on those topics, or on "culture fit" without a job-related definition. This is not legal advice.
The same principles apply when part of the interview is automated. NTRVSTA's AI interviewer, for example, produces a score breakdown with video snippets of key moments so a person can check the evidence behind each number, and its NYC Local Law 144 bias audit is published on our compliance page. Our guide on how AI interviews are scored covers what to check before trusting any automated score.
Questions
How many competencies should an interview scorecard have?
Usually four to six. OPM's structured interview guide says interviews typically assess between four and six competencies unless the job is unique or very senior. Fewer than four rarely covers the job, and more than six leaves too little time to gather evidence for each one.
Should interviewers see each other's scores before the debrief?
No. Each interviewer should submit their scores and notes before seeing anyone else's. OPM's guidance for panels is to rate independently first, then discuss differences and reach a consensus. Seeing a senior colleague's score first pulls everyone toward it.
What's the difference between an interview scorecard and an interview rubric?
The rubric is the scoring key: it describes what a 1, 2, 3 or 4 answer looks like for each competency. The scorecard is the form an interviewer fills in, with the questions, a rating for each competency, evidence notes and a recommendation. A good scorecard has the rubric built in.
Can I use the same interview scorecard for every role?
You can reuse the format, but not the content. The competencies, questions and anchors should come from the job itself, so a sales scorecard and an engineering scorecard will look very different. Shared items like communication can keep the same anchors across roles.
Need questions to put on the form? Start with our behavioral interview questions, situational interview questions or the STAR method guide. And if you'd like to see a structured first round run by an adaptive AI interviewer, book a 20-minute demo or try a sample interview first.
