Recruiting Metrics: The 5 That Matter for Small Teams
Five recruiting metrics that still mean something at under 20 hires a year, how to count each one, and how to read small numbers without fooling yourself.

The recruiting metrics that matter for a small team are time to fill, stage conversion, offer acceptance, cost per hire and 90-day retention. At under 20 hires a year, most other metrics swing so much from one hire to the next that they cannot guide a decision.
We build Talent Summoner, which searches public profiles and returns a ranked shortlist for each role, so the top of your funnel starts from a list you can count. As of 3 October 2026 your first Role is free with no card, then US$29 per Role or US$59 a month for 10 Roles (pricing).
If you hire 50 or more people a year and have a recruiting team, a full dashboard in your ATS (applicant tracking system) is worth building. If you make fewer hires, track the five below in a spreadsheet.
The 5 recruiting metrics for small teams
All examples are made up. The last column is the rule for reading a small number.
| Metric | How to count it | Made-up example | How to read it at small numbers |
|---|---|---|---|
| Time to fill | Days from the approved requisition to the accepted offer, per role | 36.5 days (median of 10 roles) | Use the median, not the average |
| Stage conversion | People who reach the next stage ÷ people who entered this stage | 5 of 20 interviewed got an offer (25%) | Compare counts, not just percentages |
| Offer acceptance | Offers accepted ÷ offers made | 8 of 10 (80%) | Read it as a range, not one number |
| Cost per hire | Total hiring spend in the period ÷ hires in the period | US$24,000 ÷ 8 hires = US$3,000 | Show the biggest single cost next to it |
| 90-day retention | Hires still employed at 90 days ÷ hires who started at least 90 days ago | 7 of 8 (88%) | One leaver is a story to read, not a trend |
How to read recruiting metrics at small numbers
Small numbers swing hard. A few habits keep you from reacting to noise.
Use the median for time to fill
One slow role can move the average a lot.
In a made-up set of 10 roles, nine took 21 to 52 days and one took 96. The average was 42 days and the median was 36.5; without the slow role they were 36 and 35.
So one role moved the average by 6 days and the median by 1.5. Report the median, and pick one start date as our guide to defining time to fill explains.
Read rates as a range
With 8 of 10 offers accepted, the rate is 80%. The 95% confidence range (Wilson method, our arithmetic) runs from about 49% to 94%.
With 40 of 50 accepted, the rate is still 80%, but the range narrows to about 67% to 89%. Our guide to confidence intervals for hiring metrics shows the method.
Fix the denominator before you compare
"Interview to offer" only means something if everyone counts the same people at the start. Write down who enters each stage, then keep that rule; our piece on defining a recruiting funnel denominator walks through it.
Small-team recruiting metrics to skip
These are common on vendor dashboards. At under 20 hires a year, they cost more effort than they return:
- Applicants per opening. It rises and falls with where you posted, not with how well you hire.
- Quality-of-hire scores. They mix manager ratings that a few hires cannot average out.
- Candidate satisfaction scores. A handful of survey replies is too few to compare one quarter with the next.
- Recruiter productivity. On a small team, the "recruiter" is often a founder with another job.
Keep the five, review them once a quarter, and write one line on what you would change.
How to run a quarterly recruiting metrics review
Put the five numbers in one spreadsheet row per role. Each quarter, answer three questions:
- Which role took longest, and at which stage did it stall?
- Where did each hire come from, and what did that source cost?
- Did anyone leave within 90 days, and what did the interview miss?
Our candidate pipeline conversion calculator helps with the first question. Our cost per hire guide covers what to count as spend.
When these recruiting metrics are wrong for you
This set is built for low volume. It is the wrong fit when:
- You hire in large batches, such as 30 sales reps a year. Then you have enough data for averages, source-quality reports and funnel benchmarks.
- You use an agency for every hire. Then the useful number is the agency fee per hire and how many of their candidates you hired.
- You have made fewer than 3 hires in total. Write notes on each hire instead of computing rates.
Where Talent Summoner fits
Talent Summoner searches LinkedIn, GitHub and other public sources, over 200 million profiles. It ranks each candidate with must-haves, nice-to-haves and plain-English reasoning.
You can save candidates and mark where each person has reached: contacted, replied, interviewed, offer or hired. Those marks give you the stage counts for the conversion metric above.
Trade-offs, stated plainly: no paid InMail, no always-on pipeline and no ATS integration today. It does not calculate time to fill or cost per hire for you.
We checked these facts on 3 October 2026 and re-check them every quarter.
What are the most important recruiting metrics?
For a small team, the most important recruiting metrics are time to fill, stage conversion, offer acceptance, cost per hire and 90-day retention. Each one can be counted in a spreadsheet and still guides a decision at under 20 hires a year.
What are recruiting KPIs?
Recruiting KPIs (key performance indicators) are the few recruiting metrics a team commits to improving, each with a target. A small team might choose a median time to fill under 45 days as one KPI.
How do you measure time to fill?
Time to fill is the number of days from the date a role is approved to the date a candidate accepts the offer. For a small team, report the median across roles, because one slow role can distort the average.
What is a good offer acceptance rate?
There is no single good offer acceptance rate for a small team, because a few offers give a wide range. Track the count, such as 8 of 10, and ask each person who declines why they said no.
Track five numbers per role, read them as ranges, and review them once a quarter.


