An interview scorecard is a structured form that lists the criteria you are hiring for and a rating scale, so each interviewer rates every candidate on the same things. Instead of leaving the room with a vague 'I liked them,' you leave with a score per criterion and a note on why. This guide covers what a scorecard is, why it beats gut-feel, how to build one, sample interview feedback comments and an example rubric you can copy.
Key takeaways
- An interview scorecard is a fixed set of criteria plus a rating scale, filled out for every candidate so you compare the same things instead of general impressions.
- Pull 4 to 6 criteria straight from the job, define what each score means in plain language, and map interview questions to the criteria they test.
- Score independently before the panel talks, then compare, so one loud voice does not pull everyone toward the same rating.
- A tight rubric makes rejections defensible and lets reviewers score the same one-way video answers without being in the room together.
What a scorecard is and why it beats gut-feel
A scorecard turns a subjective interview into structured data. Every interviewer rates the candidate on the same short list of criteria using the same scale, and writes a sentence of evidence for each rating.
Gut-feel hiring feels fast, but it rewards candidates who are similar to the interviewer and punishes anyone who interviews differently than expected. A scorecard fixes four specific problems.
- Consistency: every candidate is asked about and rated on the same criteria, not whatever came to mind that day.
- Comparability: two candidates can be lined up criterion by criterion instead of compared as vague overall impressions.
- Fairness: ratings tie to job-related evidence, which reduces the pull of personal similarity and first-impression bias.
- Defensible decisions: if a rejected candidate or your own leadership asks why, you have written, job-related reasons on file.
How to build a scorecard
Build the scorecard before you post the role, or at least before the first interview. Work through these five steps.
- 1
Define 4 to 6 criteria from the job
Read the job description and pull the handful of things that actually predict success in the role, such as core skill, problem-solving, communication, and ownership. Stop at six. More criteria feels thorough but spreads attention too thin to rate well.
- 2
Pick a rating scale
Use a short scale like 1 to 4. An even number forces a lean either positive or negative and avoids a safe middle where everyone parks.
- 3
Write what each score means
For each criterion, describe what a 1 looks like and what a 4 looks like in plain language. This shared definition is the rubric, and it is what keeps two interviewers from meaning different things by 'a 3.'
- 4
Map questions to criteria
Assign each interview question to the criterion it tests, so you know the criterion is actually being probed. If a criterion has no question behind it, add one or drop the criterion.
- 5
Score independently, then compare
Each interviewer fills out their own scorecard before the group discusses. Then compare scores and dig into the gaps. Disagreement on a criterion is a signal to review the evidence, not to average it away.
Example scorecard and rubric
Here is a sample scorecard for a mid-level role. Adapt the criteria to your job, but keep the pattern: each row names a criterion, describes what strong looks like, and carries a 1 to 4 rating.
| Criterion | What strong (a 4) looks like | Rating (1-4) |
|---|---|---|
| Core skill | Solves a realistic problem cleanly and explains the tradeoffs behind their choices. | ___ |
| Problem-solving | Breaks an ambiguous question into parts and reasons out loud instead of guessing. | ___ |
| Communication | Explains a complex idea so a non-expert follows it, and checks for understanding. | ___ |
| Ownership | Gives a concrete example of driving something end to end and owning the outcome. | ___ |
| Collaboration | Describes handling disagreement with a teammate and reaching a shared decision. | ___ |
The 'what strong looks like' column is the anchor. Writing the top of the scale down means a 4 is earned against a definition, not handed out because the candidate was likeable.
Interview feedback examples: sample comments on candidates
The comment next to each score matters more than the number. Good interview remarks quote or describe what the candidate actually said, so anyone reading the scorecard later can see why the score was given. Weak remarks describe a feeling. Here are sample interview feedback comments for common criteria, one strong and one weak.
| Criterion | Useful comment (evidence) | Weak comment (impression) |
|---|---|---|
| Core skill | Built the pivot table and lookup in the exercise without help; explained why she used a lookup instead of manual matching. | Seems technically strong. |
| Problem solving | When the first fix to the billing error failed, checked the import logs and found the date format issue. Clear steps, named the cause. | Good problem solver. |
| Communication | Explained the migration to me as a non-specialist in about two minutes, checked I followed, no jargon. | Communicates well. |
| Ownership | Raised the missed deadline with the client before their manager asked; described the new check they added afterward. | Takes ownership. |
| Collaboration | Described splitting the forecast with a peer after a disagreement; credited the peer for the regional idea. | Team player, nice person. |
| Concern | Could not give an example of handling an angry customer; both answers described what the team usually does. | Not sure about customer skills. |
- Write what was said or done, then the score. If you cannot write the evidence, mark the criterion as not covered.
- Keep comments about the job. Remarks on appearance, accent, age, family or personality traits unrelated to the work do not belong on a scorecard, and they can create legal risk.
- Note concerns as specifically as strengths, so the next interviewer knows exactly what to probe.
Tips to keep scores honest
- Avoid too many criteria. Past six, interviewers stop rating each one carefully and start assigning a single overall impression across the row.
- Avoid a 1 to 5 scale that clusters on 3. An odd scale gives a comfortable neutral middle, so ratings pile up on the center and stop separating candidates. Use 1 to 4.
- Calibrate reviewers. Before you rely on scores, have interviewers rate the same recorded or practice answer and compare, so a '3' means the same thing to everyone.
- Require a note per rating. A number with one line of evidence is far harder to fake than a number alone.
One way to remove the memory problem entirely is one-way, asynchronous video interviews, where candidates record answers to set questions. The whole panel scores the same recording against the same rubric on their own time, so nobody is reconstructing what was said, and you can rewatch a specific answer before committing to a rating.
Take the template
It is deliberately small. A scorecard that takes fifteen minutes to fill in does not get filled in, and the one that gets abandoned halfway through a hiring round is worse than none at all, because half your candidates end up scored and half remembered. Six criteria, one line of evidence each, and a decision in a sentence is the most that survives contact with a busy week.
The evidence column is the part people skip and the part that does the work. Writing what the candidate actually said, rather than a score alone, is what lets you defend a shortlist a week later when nobody remembers the interviews, and what stops a strong impression in the first two minutes from quietly setting every score that follows.
Filling it in without gaming yourself
- Score each criterion immediately after the answer, not at the end. Scores written at the end are a record of who you liked.
- Write evidence before the number. If you cannot quote anything, the score is an impression and should be marked not covered instead.
- Never adjust an early score after meeting a later candidate. Compare totals afterwards; that is what the totals are for.
- Score independently when there are two interviewers, then compare. Discussing first means the more senior person's view becomes both views.
- Treat a missing must-have as a stop, not a deduction. That is the whole point of marking it must-have.
