How to measure quality of hire: a formula you can defend, with worked examples

To measure quality of hire, pick three or four outcomes you can collect for every new hire (a manager rating at 90 days, a performance rating at 12 months, retention at 12 months, and time to full productivity), put each on a 0 to 100 scale, and average them with agreed weights. Then compare the score across cohorts, sources and interview scores. The number is only useful if you define it once and collect it the same way every time.

This page gives you the components, a 90-day survey you can copy, the formula worked through with numbers, and the step most teams skip: checking whether your interview scores predicted any of it. For agency funnel measures such as fill rate and submittal-to-interview ratio, see recruiting agency metrics.

Choose the components

Quality of hire is not one measure. It is a small set of outcomes that together answer "did we hire someone who does the job well and stays?" Choose components you can collect reliably for every hire, not the ones that sound most complete.

Three components collected for every hire beat six collected for half of them. If performance reviews at your organization do not differentiate, leave them out and use the manager survey twice, at 90 days and 12 months.

The 90-day manager survey

Five questions, a fixed scale, sent automatically on day 90. Ask about specific behavior, not overall happiness, so managers cannot simply answer "fine".

QUALITY OF HIRE CHECK: 90 DAYS
New hire: [ ]   Role: [ ]   Start date: [ ]   Manager: [ ]

Rate each 1-5:  1 = well below what the role needs
                3 = what we expect at 90 days
                5 = well above what the role needs

1. The quality of their work so far                          [ ]
2. How quickly they have become productive                   [ ]
3. How well they work with the team and stakeholders         [ ]
4. The skills we hired for are present in the actual work    [ ]
5. Knowing what you know now, would you hire them again?     [ ]

Which must-have from the hiring brief has been most visible?
[ ]
Which, if any, has been missing?
[ ]
Anything the interview process missed?
[ ]

Questions 4 and 5 and the three text boxes are the ones that improve hiring. The ratings give you a number; the text tells you which part of the interview to change.

The formula, worked through

Convert each component to a 0 to 100 scale, then apply weights that add up to 100%. The example uses invented numbers for one hire.

Converting from a 1 to 5 scale with (rating − 1) ÷ 4 puts a 1 at 0 and a 5 at 100, so a middle rating of 3 scores 50. Write the conversion rules down. If one quarter someone scores retention as 100/0 and the next as a partial credit for months served, the trend line means nothing.

Handling hires who leave early

A hire who leaves at month five has no 12-month rating. Decide in advance: score retention as 0, and either leave the performance component out and re-weight the rest, or use the last manager survey. Whichever you choose, apply it to everyone and record the reason for leaving separately.

Reading the score by cohort and source

A single hire's score is a conversation starter. The value comes from averages across groups that you can do something about:

• By source: referrals, job boards, agencies, internal moves. If one agency's placements score consistently lower, you have a specific conversation to have.
• By role family: engineering and customer support have different components and weights, so compare within a family, not across.
• By hiring manager: handle carefully. Differences may reflect how the manager rates as much as who they hire; calibrate raters before drawing conclusions.
• By quarter: did a change to the interview process move the score for the hires that came through it?

An invented example of the report for one role family over a year:

Agency B's five hires are too few for a firm conclusion, but the repeated "missing" answer is a specific point to raise at the next intake with that agency, and it points to a question your own interviews may also need to ask harder.

Keep group sizes in view. An average of four hires moves a long way when one person leaves. Show the count next to every average, and do not report a group with fewer hires than you would trust.

Check whether your interviews predicted it

This is the reason to measure quality of hire at all: to find out which parts of your process predicted the outcome, and fix the ones that did not. You need structured interview scores stored per competency. Free-text impressions cannot be compared.

1. For each hire from the last 12 to 18 months, put their interview score per competency next to their quality of hire score.
2. Sort by quality of hire. Look at the top and bottom groups. Did the top group score higher on any competency at interview?
3. Look for competencies where the scores were high for almost everyone. A competency every hire passes cannot tell good hires from weak ones; the questions or anchors need work.
4. Read the 90-day survey answers to "what did the interview process miss?" and map each to a competency or a missing question.
5. Change one thing, such as a question, an anchor or a work sample, and see whether the next cohort looks different.

With small numbers, this is structured judgment rather than statistics, and that is fine. The interview scorecard template gives you per-competency scores in a comparable format, and scorecards that carry the candidate's own words make step 4 much easier. Interview Signal attaches quotes checked against the transcript to each score, which is the kind of record this review needs.

Quality of hire for agencies

Agencies rarely see a client's performance reviews, but they can build a useful version from what they do see:

The 90-day client call doubles as account management. Ask the two questions, then ask what else the team is hiring for.

Mistakes that make the number meaningless

Setting it up in a month

• Week 1: agree components, weights and conversion rules with HR and two hiring managers. Write them on one page.
• Week 2: set the 90-day survey to send automatically from your HR system or a calendar reminder, and add a "productivity milestone" field to the intake form.
• Week 3: backfill the last 12 months of hires where the data exists, and mark missing data rather than guessing.
• Week 4: produce the first report by source and role family, with counts, and take one finding to the next calibration session.