The recruiting metrics worth tracking, and how to read them
Recruiting metrics are the numbers that describe how hiring is going, from pipeline volume through to how the hire worked out. Recruitment metrics and recruitment KPIs name the same arithmetic. Start with 4: time to fill, screen to interview conversion, offer acceptance rate and first-year attrition. Add the rest once those are honest.
Settle this before anyone reports a number
Time to fill counts from the requisition being approved. Time to hire metrics count from first contact with the person you eventually hired. Both conventions are in live use, which is how two teams compare unlike numbers and argue.
8 recruiting metrics, with the formula and the failure mode
This is the reference, and the recruiting metrics examples most teams start from. Work the middle column first: a metric nobody can compute the same way twice is not a measurement yet.
| Metric | How to calculate it | What it tells you | Where it usually breaks |
|---|---|---|---|
| Time to fill | Days from requisition approval to offer acceptance. | The wait the business actually experiences, approvals and posting included. | Blamed on the recruiter when half the days sit in approval and scheduling. |
| Time to hire | Days from first contact with the eventual hire to their acceptance. | How fast your process moves once a real candidate is in it. | Reported next to somebody else's time to fill as though they match. |
| Screen to interview conversion | Candidates interviewed divided by candidates screened, times 100. | Whether your screen is too loose, too tight, or applied inconsistently. | Nobody records who was screened, so the denominator is a guess. |
| Interview to offer ratio | Candidates interviewed divided by offers made, stated as a ratio. | How much panel time one offer costs you. | Quoted as a ratio by one team and a percentage by another. |
| Offer acceptance rate | Offers accepted divided by offers extended, times 100. | Whether the role, the pay and the process match what was described. | Computed on 5 offers, then treated as a trend. |
| Cost per hire | Total recruiting costs in a period divided by hires in that period. | What a hire costs you once agency, advertising, tooling and time are counted. | What counts as a cost changes between quarters, so the trend is meaningless. |
| Quality of hire | No single agreed formula. A common proxy blends first-year retention with a manager rating at 6 or 12 months. | Whether speed is producing hires who work out. | Two people quote it with different arithmetic and nobody notices. |
| First-year attrition | New hires who left within 12 months divided by total new hires, times 100. | The lagging check on every other number above. | Read alone, so voluntary and involuntary exits get mixed together. |
Quality of hire genuinely has no standard formula. Write down the one you use and hold it still for a year.
4 more hiring metrics to add once those 8 are honest
- Qualified candidates per opening: pipeline coverage, and it warns you weeks before time to fill does
- Source of hire and sourcing channel efficiency: hires per channel, and cost per hire split by channel
- Response rate on outbound: replies divided by people contacted, the one recruiter metric sourcers own
- Hiring manager satisfaction: a quarterly rating, qualitative, and the one that decides whether anyone trusts the rest
Is a recruitment KPI different from a metric?
Recruiting KPI A recruiting metric with a target and an owner attached. The arithmetic does not change. Someone has agreed what good looks like and whose job it is to move it.
So recruitment KPIs, hiring KPIs, KPIs for recruiters and any KPI for recruitment you have been told to set are the same numbers under different job titles. Track 12, promote 4 with a target. The full set by stage is in our recruitment KPIs guide.
Recruitment analytics in 3 stages, from counting to predicting
Recruitment data analytics is not one capability. Data analytics in recruitment arrives in an order, and skipping a stage is why dashboards get abandoned.
- Count Get the stage counts honest before anything else. If nobody records who was screened or when a candidate entered a stage, every rate above it is fiction dressed as analytics. 2 dates per stage per candidate. That is the whole requirement.
- Compare Segment before you compare. Healthcare recruiting metrics and campus recruiting metrics are not the same funnel, and neither is a backfill and a first hire in a new country. One blended average hides every problem in it. Split by role family, seniority, location and channel. Then compare like with like.
- Predict Predictive analytics in recruitment needs repeat volume of the same role and clean historical outcomes. Most teams have neither, and forecasting from 9 hires produces confident nonsense. If you fill the same role 40 times a year, forecasting is reasonable. If you fill it twice, it is not.
What data driven recruitment actually means
There are 2 kinds of recruitment data. The first you already hold: your ATS timestamps, your outreach replies, your interview scorecards, your offers. The second you buy, such as market salary sets or talent pool sizing. Data driven hiring almost always fails on the first kind, because nobody recorded it consistently, not because the second was unavailable.
Data driven recruiting means changing a decision because of a number. If recruitment data analysis goes on a slide and the same 3 roles get worked the same way next quarter, that is reporting, not data driven recruitment. The test is simple: name the decision each metric is allowed to change, in advance, and what value would trigger it.
Build a recruitment KPIs dashboard in one afternoon
A recruitment dashboard fails from ambition, not from tooling. 7 decisions, then whatever spreadsheet or report builder you already have.
- Pick 4 metrics, not 14. You can add more after 2 quarters of them being right
- Write the definition of each one in a sentence, next to the number itself
- Name the single system of record each number is pulled from
- Set the cadence: pipeline weekly, quality and cost quarterly
- Baseline against your own trailing 12 months, not a published average
- Segment by role family before anything is charted
- Decide in advance what action each number triggers, and who takes it
Where the stage numbers come from
Candidates grouped by stage, each group carrying its own status and count, with bulk invite on the header.
On The Cognitive the pool starts with AI sourcing across ~900M profiles into a durable shortlist. Everyone who enters a role is then tracked by stage, with explicit statuses and a dated history per candidate. Recruitment monitoring and recruiter metrics are a byproduct: the counts exist because the work happened in one place.
Reading a recruiter performance metric without fooling yourself
Do
- Baseline against your own history, over at least 3 consecutive periods
- Segment by role family and seniority before you compare anything
- Pair every speed metric with a quality metric so nobody games the clock
- Judge recruitment success metrics on the trend, not on this month's value
- Say out loud how many events a rate is computed from, every time you show it
Do not
- Comparing your blended average to somebody else's blended average
- Reporting a rate computed on fewer than about 20 events
- Trying to optimize recruitment metrics one at a time: time to fill alone lowers the bar
- Changing a definition mid-quarter and keeping the old chart
- Holding a recruiter to performance metrics for recruiters they cannot influence
Why recruiting metrics benchmarks are worth less than your own baseline
Most benchmark figures in this field arrive with no citation, or with a citation to a report sitting behind a paywall. Then they get quoted onward until they read as common knowledge. Role, seniority, industry and location move these numbers further than anything a hiring team controls, so a benchmark from outside your segment is worse than no benchmark at all.
The market moves too. BLS reported 7.3 million job openings and 5.1 million hires in July 2026, in a release published on 1 September 2026, so the conditions a figure was measured in may not be the conditions you are hiring in. Compute your own trailing 12 months, per role family, and beat that. It is the only comparison that fits your roles.
Response rate is a number you already have
Replies sorted by intent with the interested ones first, and a count for each intent above the table.
Outbound response rate is replies divided by people contacted, and it is the one recruiter analytics figure most teams never pull. Outreach runs as email from your own mailbox on a Day 0, plus 3, plus 4 cadence, and stops the moment someone replies. Each reply is classified by intent.
A weekly recruiting report anyone can read
Most talent acquisition reporting and analytics fails by being long. This fits on one screen and takes 10 minutes to fill in from your own records.
WEEK OF: [date] · PREPARED BY: [name] 1. ROLES OPEN Role · days open · stage the role is sitting in · owner Flag any role open longer than your own trailing median for that role family. 2. PIPELINE COVERAGE Role · qualified candidates in play · target coverage · gap One line per role. A coverage gap this week is a time to fill problem in 3 weeks. 3. WHAT MOVED THIS WEEK Entered screen: n · Entered interview: n · Offers out: n · Accepted: n · Declined: n Rejections and withdrawals by stage, so drop-off has a location. 4. THE ONE NUMBER THAT CHANGED Metric · last period · this period · how many events it is computed from Only include it if it moved more than normal week to week variation. 5. WHAT WE ARE DOING ABOUT IT The decision, the owner, the date we check again. If there is no decision here, delete section 4 rather than reporting a number nobody acts on. 6. ASKS Named person · what is needed · by when. Usually interview slots or a pay range. QUARTERLY ADDENDUM (not weekly) Cost per hire by channel · quality of hire proxy · first-year attrition · hiring manager rating.
Diversity recruiting metrics are selection rates, and US employers keep them
In the US, the Uniform Guidelines at 29 CFR 1607.4(A) say each user "should maintain and have available for inspection records or other information which will disclose the impact which its tests and other selection procedures have upon employment opportunities of persons by identifiable race, sex, or ethnic group". The EEOC points employers to that Part for whether a selection procedure is lawful.
The benchmark is at 1607.4(D): a selection rate for any race, sex or ethnic group below four fifths, or 80 percent, of the rate for the highest group "will generally be regarded by the Federal enforcement agencies as evidence of adverse impact". So the useful diversity metric is a pass rate at each stage, compared across groups, not a headcount pie chart. Check local law elsewhere.
A number nobody has agreed the definition of is not a measurement. It is an argument waiting for a meeting.
Go deeper on one metric
What is cost per hire? · What is quality of hire? · What is cost per hire? · 14 recruitment KPIs by funnel stage · The recruitment process in 7 steps · Data-driven recruitment
Frequently asked questions
What is recruitment analytics?
Recruitment analytics is using your own hiring data to answer a specific question and change a decision. It arrives in 3 stages: counting, where stage records are made honest; comparing, where numbers are segmented by role family and seniority; and predicting, which needs repeat volume of the same role. Most of the value in recruiting analytics sits in the first stage.
What are the most important recruiting metrics to track?
Start with 4: time to fill, screen to interview conversion, offer acceptance rate and first-year attrition. Those cover speed, the screen, the close and whether the hire worked out. 4 is deliberate. A team tracking 14 key recruitment metrics from a standing start usually has 14 numbers nobody trusts and no agreed definitions.
What is the difference between time to fill and time to hire?
Time to fill counts from the requisition being approved to the offer being accepted, so it includes approval and posting delays. Time to hire counts from first contact with the person you eventually hired to their acceptance, so it measures your process. Both conventions are genuinely in use. Write down which one you mean before anyone reports it.
What is the difference between a recruiting metric and a recruiting KPI?
None arithmetically. A KPI is a metric with a target and a named owner attached. Recruitment KPIs, recruiting KPIs and talent acquisition KPIs describe the same calculations you already run, promoted to something a team is accountable for. A practical split is 12 metrics tracked, 4 of them promoted to KPIs with a target.
How do you calculate cost per hire?
Divide total recruiting costs in a period by the number of hires in that period. The arithmetic is trivial; the agreement is not. Decide whether recruiter salaries, agency fees, advertising, tooling, referral bonuses and relocation are in or out, write it down, and hold it constant. A cost per hire trend only means something if the inputs never changed.
What are good recruiting metrics benchmarks?
Published recruiting metrics benchmarks vary enormously by role, seniority, industry and location, and the credible sources are mostly paywalled while the widely repeated figures usually arrive with no citation at all. Your own trailing 12 months, split by role family, is a better comparison than any national average. Compute it first, then treat outside figures as context only.
How do you build a recruitment KPIs dashboard?
Pick 4 metrics. Write the definition of each next to the number. Name the single system of record each is pulled from. Set 2 cadences, weekly for pipeline and quarterly for quality and cost. Baseline against your own trailing 12 months. Then decide, before anyone looks at it, what action each number is allowed to trigger.
Does predictive analytics in recruitment work?
It works where you fill the same role repeatedly and have clean historical outcomes to learn from. High-volume and campus hiring qualify. Most teams do not, and forecasting from a handful of hires produces confident output with nothing behind it. Predictive hiring analytics and talent acquisition predictive analytics sit at stage 3 of the ladder. Clean stage counts return more for almost everyone.
What KPIs should a talent acquisition manager set?
The best KPIs for talent acquisition managers describe the function, not the individual: time to fill by role family, offer acceptance rate, first-year attrition and hiring manager satisfaction. Add pipeline coverage per open role as the leading indicator. Avoid any KPI for talent acquisition built on numbers they cannot influence, such as approval delays or a pay band set elsewhere.
How do you measure diversity recruiting metrics?
Measure pass rates at each stage and compare them across groups, rather than reporting headcount shares. In the US, 29 CFR 1607.4 asks employers to keep records that disclose the impact of their selection procedures by race, sex and ethnic group, and treats a selection rate below four fifths of the highest group's rate as evidence of adverse impact. Rules differ outside the US.
What is a good interview to hire ratio?
Interview to hire ratio is candidates interviewed divided by hires, and the right value depends on the role, so compare against your own history for that role family. Read the direction. A ratio climbing over several quarters usually means the screen before interviews is letting through people it should not. A ratio that is very low can mean you filter so hard before interviewing that good candidates never get a conversation.
What is candidate drop-off rate?
Candidate drop-off rate is the share of people who leave a stage without completing it: candidates who exit a stage divided by candidates who entered it, times 100. Compute it per stage, never once for the whole funnel, so the loss has a location. Drop-off that clusters around waiting, for a screen or an interview slot, is a capacity problem rather than a lack of candidate interest.
Explore: The recruitment process in 7 steps · What is cost per hire? · What is quality of hire? · Candidate experience · The recruitment funnel and drop-off rate · Recruitment funnel conversion rates · Recruitment analytics software · Employee turnover rates by industry