Candidate scorecard: a free template and how to use it

A candidate scorecard lists the job’s requirements as rows, rates each candidate against each one from evidence you actually saw or heard, and combines those ratings by a rule you set before meeting anyone. Its job is to stop the decision being made by whoever interviewed last, or best.

Download the template: printable PDF · spreadsheet (CSV). Free, no sign-up.

The rest of this guide shows how to fill it in, how to weight it, and how to turn it into a decision.

Why a scorecard helps

Without one, candidates are compared on overall impressions formed at different times, by different people, in different moods. The most recent interview is the most vivid, and a confident manner can stand in for evidence.

The case for scoring is stronger than “it’s tidier”. A meta-analysis by Kuncel and colleagues (2013) found that combining information about candidates mechanically, by formula or fixed rule, predicted job performance substantially better than combining it by holistic judgement, even when the people judging were experts in the job. A scorecard is the simplest mechanical method there is.

What goes on it

Each row is one thing the job needs. Each candidate gets three entries per row.

Column	What goes in it
Requirement	One job requirement, written as something you can see evidence of. Your screening must-haves, plus what the interview tests.
Rating	1 to 5, using anchors you wrote in advance (what a 1, 3 and 5 look like for this row).
Evidence	What you saw or heard that justifies the rating: a line from the application, an interview answer, a work sample result.
Unclear	A tick if you don’t have enough evidence to rate it yet. Kept separate from a low rating on purpose.

The rows come from the job, not from a generic list. If you followed our guide to screening resumes you already have them: the must-haves and nice-to-haves you screened on. Add a row for anything the interview is there to test, such as how someone reviews other people’s work.

Leave out rows you can’t rate from evidence. “Culture fit”, “presence” and “energy” are usually impressions with a label. If something genuinely matters, write down what you would see from someone who has it, and make that the row.

A filled-in example

Here is a scorecard for two finalists for the fictional Senior Accountant role at Larkspur Partners used throughout our guides, after a structured interview. The candidates and answers are invented to show how the scorecard works.

Requirement	Candidate A	Evidence (A)	Candidate B	Evidence (B)
CPA designation	5	Designation listed; confirmed in screening question	5	Designation listed; confirmed in screening question
Runs month-end close independently	5	Described owning the close calendar and sign-off at a 40-person firm (interview Q1)	3	Owned reconciliations and accruals; controller coordinated and signed off (Q1)
Handles a close problem under time	3	Sensible order of steps; didn’t mention telling the partner early (Q2)	5	Isolated the reconciling item first, flagged the partner the same afternoon (Q2)
Prepares audit working papers	4	Prepared lead schedules for two audits; changed process after auditor requests (Q3)	Unclear	Mentioned “supporting the audit” but couldn’t say which files they prepared (Q3)
Learns new systems	4	Led part of a migration to a cloud ledger (Q4)	4	Learned two systems in three years; built a checklist for the team (Q4)
Reviews others’ work constructively	3	Would raise it with the manager first (Q5)	4	Would show the junior the pattern with an example, then tell the manager (Q5)

Notice what the scorecard does here. Candidate B’s audit row is marked unclear, not scored 1, because “supporting the audit” is not evidence either way. That is a question to settle with a short follow-up call, not a point to deduct.

Weighting: start with equal weights

The tempting move is to give each row a carefully considered weight: 30% for this, 15% for that. The evidence suggests you don’t need to.

A review and meta-analysis by Bobko, Roth and Buster (2007) found that equal weights “have substantial predictive validity” compared with statistically derived weights, and concluded they “can be a highly appropriate approach for weighting under many circumstances”. The U.S. Office of Personnel Management recommends equal weights for structured interview competencies, and says that if you weight them differently, you should document why.

So: weight every row equally unless you have a specific reason not to, and if you do, write the reason down before you see any candidates. A simple, common exception is counting must-haves double. What you should never do is change the weights after the ratings are in. At that point you are no longer weighting the job; you are adjusting the result toward the candidate you already prefer.

From ratings to a decision

1. Each interviewer fills in their own scorecard first, alone, straight after the interview. Comparing notes before rating pulls everyone toward the most confident voice in the room.
2. Resolve the unclears. For each unclear row on a finalist, either get the missing evidence (a five-minute call is often enough) or agree how you’ll treat it, for example by leaving it out of that candidate’s average and noting it.
3. Combine by your rule. Average each candidate’s ratings across rows, and across interviewers if there are several. To put every finalist’s evidence from every stage side by side, see how to compare candidates, which has a free comparison matrix.
4. Then discuss. The discussion is for catching what someone missed or misheard. Where interviewers differ by more than a point on a row, go back to the evidence written in that row, not to anyone’s overall impression.
5. Record the decision and the reason. One or two sentences, tied to the rows. It is the cheapest way to learn from the hire later.

In the example, with equal weights and Candidate B’s audit row left out pending a follow-up, Candidate A averages 4.0 across six rows and Candidate B 4.2 across five. That is close enough that the follow-up on B’s audit experience could decide it. A good scorecard often does exactly this: it narrows the decision to the one question that actually separates the candidates.

Mistakes that undo the scorecard

• Rating the person, not the row. If every rating for a candidate is the same number, the scorecard was filled in from an overall impression.
• Rows that measure polish. “Communicates clearly” can end up rewarding interview confidence. Tie it to the job: “explains a reconciling item so a partner can act on it”.
• Scoring “unclear” as low. It punishes candidates for what you didn’t ask.
• Filling it in days later. Memory fills gaps with impressions. Rate the same day, from notes.

What a scorecard won’t do

A scorecard makes your evaluation consistent and explainable. It does not make it valid. If the rows are the wrong requirements, or the evidence behind them is weak, you will be consistently wrong. The strongest inputs are the ones the research supports: a structured interview, a realistic sample of the work, and evidence of job knowledge. See what actually predicts job performance.

Using the template

• The PDF prints as one page per candidate, with space for eight rows, ratings, evidence and the unclear mark, plus a decision box.
• The CSV opens in Excel, Google Sheets or Numbers. It has the example rows above, with anchors, as a starting point, and an average row that ignores blank (unclear) ratings. Replace the rows with your own requirements, and copy a candidate’s three columns to add another candidate.

Where SortTheStack fits

SortTheStack doesn’t produce interview scorecards. It covers the step that creates the first rows: it drafts requirements from your job description for you to edit, and reads every application against them, quoting the evidence it found and marking what it couldn’t find as unclear. That evidence is what you carry into the screening rows of the scorecard. See how the screening works.

Sources

• Kuncel, N. R., Klieger, D. M., Connelly, B. S., & Ones, D. S. (2013). Mechanical versus clinical data combination in selection and admissions decisions: A meta-analysis. Journal of Applied Psychology, 98(6), 1060–1072. doi.org/10.1037/a0034156
• Bobko, P., Roth, P. L., & Buster, M. A. (2007). The usefulness of unit weights in creating composite scores: A literature review, application to content validity, and meta-analysis. Organizational Research Methods, 10(4), 689–709. doi.org/10.1177/1094428106294734
• U.S. Office of Personnel Management, How do I score a structured interview?, accessed 9 October 2026.