What an interview scorecard is
An interview scorecard is a structured form every interviewer fills in for every candidate, using the same competencies, the same rating scale, and the same evidence requirement. Instead of a debrief where one person says "strong yes" and another says "something felt off," you get seven scores, a quote behind each one, and a recommendation you can defend six months later.
The format matters less than the discipline. What makes a scorecard work is that it is filled in the same way for candidate one and candidate nine, it is written down before the panel talks to each other, and every score points at something the candidate actually said. This template enforces all three.
What goes on the scorecard
The template above is split into seven sections. Each one exists because a hiring decision falls apart without it.
Candidate information
Candidate, role, interview stage, date, interviewer, current employer. Boring, and the reason you can still compare two scorecards after thirty interviews. Stage matters most: a 4 from a screening call and a 4 from a final panel are not the same number.
Hiring facts
Salary expectation, notice period, work authorization, preferred work mode, earliest start date, years of experience. These are the details that kill offers at the last minute when nobody wrote them down. Capture them in the interview, not in a follow-up email three weeks later.
Competency scores
Communication, problem solving, technical or functional skill, collaboration, ownership, motivation, and values or culture add, each scored 1 to 5 with an evidence box next to it. The evidence box is not optional. A score with no quote behind it is a feeling, and feelings are where bias enters the process.
Structured interview evidence
The questions you asked, what the candidate answered, and your read on the answer, kept separate on purpose. Mixing the answer and the assessment into one paragraph is how a panel ends up arguing about what was said instead of what it meant.
Strengths, concerns, recommendation
Strengths and concerns in plain language, then a recommendation on a five-point scale (strong hire, hire, lean hire, lean no hire, no hire) plus a confidence level. Forcing a lean rather than allowing a neutral middle is deliberate: a scorecard that always lands on "maybe" is not helping anyone decide.
The 1 to 5 rating scale, defined
A scale only produces comparable numbers if everyone reads it the same way. Anchor it before the panel starts:
- 1, Poor. Clear gap against the requirement. Would need significant support to do the job.
- 2, Below. Some relevant capability, but under the bar for this role at this level.
- 3, Meets. Does the job. This is a pass, not a disappointment, and most scores on a healthy scorecard sit here.
- 4, Strong. Above the bar. Would raise the average of the team on this dimension.
- 5, Exceptional. Top of the market. Rare, and it should feel rare when you give it.
The most common failure is scale drift, where 3 quietly becomes a failing grade and every hire needs straight 4s. If nobody on your last ten scorecards scored a 3 and got hired, your scale has drifted.
Which competencies to score
The seven competencies in this template are a starting point, not a law. Rename them to match the role. A good rule: score no more than seven dimensions, and make every one of them something the interview can actually observe.
- Sales and commercial roles: discovery quality, objection handling, pipeline discipline, commercial judgment, resilience.
- Engineering: technical depth, system design, debugging and problem solving, code collaboration, delivery ownership.
- Operations and support: process thinking, prioritization under load, customer empathy, written communication, escalation judgment.
- Leadership and executive: strategic clarity, hiring and developing people, cross-functional influence, decision quality under ambiguity, operating cadence.
Pull these from the role brief rather than inventing them per interview. If the brief does not say what good looks like, fix that first with a job brief and a must-have versus nice-to-have pass, then build the scorecard from it.
How to run the scorecard
- Agree the competencies and the scale with the panel before the first interview, not after the first disagreement.
- Fill the candidate information and hiring facts early in the call while they are still easy to ask about.
- Take evidence during the interview, not scores. Write down what was said. Scoring while someone is still talking is how first impressions become the verdict.
- Score within ten minutes of the call ending, before you read anyone else's scorecard. Independent scores first, discussion second.
- Bring the completed scorecard to the debrief. Discuss the gaps between scorers, not the vibes.
Five mistakes that make a scorecard useless
- Scoring from memory. The evidence box stays empty, and two days later the score is all that survives.
- Scoring after the debrief. Once the loudest voice in the room has spoken, everyone else's scorecard converges on it.
- Too many competencies. Fifteen dimensions means twelve of them get a lazy 3.
- Different competencies per candidate. Then the scorecards are not comparable, which was the entire point.
- No recommendation. A page of numbers with no call at the end pushes the decision back into the meeting it was supposed to settle.
Scorecard, evaluation form, or assessment sheet?
These names get used interchangeably. An interview evaluation form and a candidate assessment scorecard are the same artifact: a per-interview record of scores, evidence, and a recommendation. What varies is how much structure the form imposes. A form that asks only for free-text impressions gives you nothing to compare. A scorecard with fixed competencies, a fixed scale, and mandatory evidence gives you a hiring decision.
Filling it from the interview instead of from memory
The hard part of any scorecard is not the format, it is the twenty minutes of writing after a call you already spent an hour on. That is why scorecards get skipped and why evidence boxes end up empty.
Sonarnote records the interview without a bot joining the call, produces a speaker-labeled transcript, and drafts the hiring facts, the answer summaries, the competency scores, and the recommendation from what was actually said. You review and correct instead of writing from a blank page, and every score keeps the quote that produced it.