A fit score is a single number that shows how closely a candidate, company, or deal matches a defined bar, calculated by rating each criterion in a rubric, weighting the ratings, and adding them up, instead of relying on one reviewer's overall impression.
How a fit score is calculated
Most fit scores are a weighted sum. Robyn Dawes showed in 1979 that even simple weighted sums of the right criteria tend to beat unaided expert judgment. A 2020 test on three hiring datasets by Yu and Kuncel in Personnel Assessment and Decisions went further: consistent random weights reliably beat the experts. Each criterion gets a rating on a common scale and a weight that reflects how much it matters. The formula is:
fit score = (w1 x r1 + w2 x r2 + ... + wn x rn) / (sum of weights x top rating) x 100
Dividing by the maximum possible total turns the result into a 0 to 100 number, which is easier to read across roles or deal types. The steps:
- List the criteria that define the bar, usually 4 to 8. More than that and the weights stop meaning much.
- Set a weight for each, adding to 100 or to 1.
- Rate the candidate or company on each criterion, with a reason.
- Multiply, add, and divide by the maximum.
- Keep the per-criterion ratings next to the total. The total ranks; the parts explain.
The weighted sum is the simplest member of a larger family of methods called multiple-criteria decision analysis. It assumes criteria can trade off against each other: a high rating on one can make up for a low rating on another. Where that is not acceptable, use a pass or fail check first (see knockout criteria).
Why teams use fit scores
A single number lets a team sort a long list into a ranked shortlist in seconds and talk about the same thing. "She is a 78 on this role" is a starting point for a debate. "I liked her" is not.
The idea of fit comes from organizational psychology, where person-environment fit describes how well a person's traits match a job, team, or organization. A 2023 review by Kristof-Brown, Schneider and Su in Personnel Psychology looks back on 50 years of this research and concludes that fit with the organization matters quite a bit, above all for how people feel about their work and whether they leave. Investors use the same logic for companies against a mandate. Sales teams use it for accounts against an ideal customer profile. In every case, the score is only as good as the definition of the bar behind it.
Worked example: a fit score for a head of finance
Pinecrest Health, a fictional 80-person company, is hiring a head of finance. The hiring manager sets five criteria, rated 1 to 5:
| Criterion | Weight | Candidate A | Candidate C |
|---|---|---|---|
| Has closed a Series B or later round | 30 | 5 | 3 |
| Built a finance team from 1 to 5 people | 25 | 3 | 5 |
| Healthcare billing experience | 20 | 2 | 5 |
| Board reporting | 15 | 5 | 3 |
| Systems migration | 10 | 4 | 4 |
| Fit score (0 to 100) | 76 | 80 |
Candidate A totals 380 out of a possible 500. Candidate C totals 400. The totals are close, but the profiles differ sharply: A is a fundraiser, C is a builder who knows the industry. The fit score puts C slightly ahead, and the per-criterion ratings tell the hiring manager exactly what she would be trading. This is the same arithmetic behind job fit scoring, laid out on a scorecard.
Fit score vs thesis fit, job fit and ICP fit
"Fit score" is the general term. The specific versions differ in what the bar is and who sets it.
| Term | Scores | Against | Set by |
|---|---|---|---|
| Fit score | Anything | Any defined bar | Whoever owns the decision |
| Thesis fit | Companies | A fund's thesis and mandate | Partners |
| Job fit | Candidates | A role's requirements | Hiring manager |
| ICP fit | Accounts | The ideal customer profile | Sales and marketing leadership |
Common mistakes
- Hiding the parts. A total of 72 with no breakdown cannot be challenged or explained.
- Weights nobody agreed to. If partners would set the weights differently, the score settles nothing.
- Letting a high score override a must-have. A weighted sum will happily rank a candidate without the required license above one with it.
- Comparing scores across different rubrics. A 75 on one role's scorecard and a 75 on another's are not the same thing.
- Treating the cutoff as fixed. Look at past hires or investments and see where the good ones landed before picking a threshold.
How ScoringFactory builds fit scores
ScoringFactory learns a team's bar from the decisions it has already made, then scores each founder, company, or candidate against it. Each score shows the criteria behind it and the record each rating came from, so the team can see what drove a 76 versus an 80 and make the call. See how it works.
Frequently asked questions
What is a fit score?
A fit score is a single number, often 0 to 100, that summarizes how well a candidate, company, or account matches a defined set of criteria. Each criterion is rated, weighted by importance, and added up. The breakdown by criterion should always travel with the total, because the total alone cannot be explained or challenged.
How is a fit score calculated?
Rate each criterion on a common scale, multiply each rating by its weight, add the results, and divide by the highest possible total. Multiply by 100 for a percentage. For example, a weighted total of 380 out of a possible 500 gives a fit score of 76. Apply any must-have checks before the weighted sum, not inside it.
What is a good fit score?
There is no universal number. A good score is one that past successful hires or investments tended to reach on the same rubric. Score 10 to 20 past cases you know turned out well and see where they cluster. Use that range as the starting threshold, then adjust as new outcomes arrive.
Should a fit score decide who moves forward?
No. A fit score is a sorting and discussion tool. It tells a team where to look first and what to talk about. Someone still needs to read the reasons behind each rating, consider what the rubric does not capture, and make the decision. In hiring, automated scores used to screen people may also carry legal obligations.
Sources
- Yu and Kuncel (2020), Pushing the limits for judgmental consistency: comparing random weighting schemes with expert judgments, Personnel Assessment and Decisions
- Kristof-Brown, Schneider and Su (2023), Person-organization fit theory and research: Conundrums, conclusions, and calls to action, Personnel Psychology (Wiley)