Interview Scorecard Template: How to Score Candidates 1 to 5

An interview scorecard is a short, written form that lists the competencies a role requires, defines what a weak and a strong answer looks like for each one, and gives every interviewer the same 1 to 5 scale to score against. It exists so that a hiring decision comes from evidence rather than from who felt like a good fit. This guide gives you a scorecard template you can copy, the anchor wording that makes the scores mean something, and the legal line about what must never appear on one.

What is an interview scorecard?

It is the difference between "I liked her" and "she scored a 4 on incident handling, because she walked through the outage she led and named the tradeoff she got wrong." Both are opinions. Only one of them can be compared to another candidate, explained to a hiring manager, or defended if the decision is ever questioned.

A scorecard turns the interview into a measurement. You decide before you meet anyone what you are measuring, what good looks like, and how you will score it. Then every candidate gets the same questions and the same scale, and the notes go in the same boxes. It sounds bureaucratic. In practice it saves time, because the argument about who to hire stops being a debate about vibes.

Scorecard, rubric, or evaluation form?

The words get used interchangeably, and the distinction is worth thirty seconds. The rubric is the scoring key: the competencies and the definition of each score level. The scorecard is the form the interviewer fills in for one candidate, using that rubric. An evaluation form is the loose version most companies actually use, which collects impressions without defining what any score means. The evaluation form is the one to abandon.

The interview scorecard template

Copy this into a spreadsheet, one column per candidate. Six competencies is a good target and twelve is too many, because scores stop being considered once the form gets long.

The evidence column is the one people skip and the one that does all the work. A score with no quote behind it is just a number someone felt like writing, and it will not survive a calibration meeting or a challenge from a candidate.

Write anchors, not adjectives

A 1 to 5 scale labeled poor, fair, good, very good, excellent is useless, because nobody agrees what good means. Anchored scales describe observable behavior at each level. Here is the same competency, done badly and done properly.

Weak version: Problem solving. 1 = poor, 3 = average, 5 = excellent.

Anchored version: Problem solving.

• 1. Describes the problem only. No structure, no options considered, no outcome.
• 2. Names one approach. Cannot explain why it was chosen over anything else.
• 3. Explains the approach and the outcome. Tradeoffs are mentioned only when prompted.
• 4. Lays out the constraint, the options weighed, the choice made, and the result, unprompted.
• 5. All of the above, plus what they would do differently now and what it taught them.

Write that once per competency and every interviewer on your panel is suddenly scoring the same thing. This is also the single highest-leverage fix for panel disagreement, which is almost never a disagreement about the candidate. It is a disagreement about the scale.

How do you weight the competencies?

Ask what actually predicts performance in the role, not what is nice to have. For a support hire, communication and composure carry most of the weight and technical depth carries little. For a controller, the reverse. Set the weights before you interview anyone, because weights chosen after you meet a candidate you like are not weights, they are a justification.

Keep the math simple. Score each competency 1 to 5, multiply by its weight, add them up. A candidate at 4.2 beats a candidate at 3.6, and if the gap is inside a tenth of a point, the honest conclusion is that your scorecard cannot tell them apart and you need a different signal, not a longer argument.

What must never go on a scorecard

This is the part most templates skip. A scorecard is a written record, and in the United States a written record is discoverable. Anything on it that is not job-related is a liability, and it is also just bad hiring.

• No notes on age, race, national origin, religion, disability, pregnancy, marital or family status, or accent.
• No "culture fit" as a scored line. It is unmeasurable and it is where bias goes to hide. If you mean something specific, like works well without supervision, score that instead.
• No comments on appearance, home background visible on a video call, or how someone sounds.
• No scores for anything the candidate was never asked about.

Everything on the form should trace back to a question every candidate was asked. That is the whole test, and it is the same standard EEOC guidance applies to any selection procedure.

Calibrate before the first candidate, not after the last

Get your panel in a room for twenty minutes. Take one real answer, ideally a mediocre one, and have everyone score it independently against the anchors. You will find a spread of two points, and you will find out that half the panel thinks a 3 is a pass and the other half thinks it is a rejection. Fix that before you interview anyone, not in the meeting where you are deciding on a real person.

Interviewers do not disagree about candidates nearly as often as they disagree about what the numbers mean. Calibration is the cheapest quality improvement available to a hiring team.

When does the scorecard get filled in?

During the interview if you can, and within ten minutes of it ending if you cannot. Memory decays fast and it decays selectively, which means a scorecard written the next morning records your impression rather than the candidate's answers. If you are filling in five scorecards on Friday for interviews you ran on Tuesday, you are not scoring candidates, you are ranking your memories of them.

This is also the practical reason structured screening at volume is so hard to do by hand. The discipline is not complicated, it is just relentless, and it is the first thing to slip when a recruiter has nine screens booked and a hiring manager asking for the shortlist by lunch.

Scoring the first round without doing it yourself

Everything above describes what a good first-round screen looks like: the same questions, the same anchors, the same scale, evidence quoted, filled in immediately. That is exactly the work our AI recruiter does on every applicant. The agent asks your structured questions, follows up when an answer stops short of the anchor, scores each competency against the rubric you defined, and quotes the transcript line behind every score. Your recruiters open a ranked shortlist and make the call.

The scoring approach is covered in more depth on our candidate interview scoring page, and if you are still deciding whether to structure the first round at all, the case for it is in our piece on structured versus unstructured interviews. Once someone is hired, the same competency list is the natural backbone for their first ninety days, so it is worth carrying the scorecard straight into whatever you use to onboard and train the new hire rather than starting from a blank page.