Hiring Playbook

What are recruiting KPIs? How to set them, the list and examples

A guide for SMBs: how to set recruiting KPIs, with benchmarks and fixes for 10 metrics, and examples from mid-career, new graduate and engineer hiring.

Illustration of a chart and a checklist working back from a hiring target to applications, interviews and offers
Emmanuel Gendre Former Google recruiter, HR and hiring advisor to SMBs Last updated 32 min read

Basics

What are recruiting KPIs? The targets for each stage before a hire

In short

Recruiting KPIs are the targets you set for each stage before a hire (applications, interviews, offers, the offer acceptance rate) to reach your hiring goal, or KGI: how many people, of what profile, by when. You set them by working back from the number of hires, one stage at a time, using each stage's pass rate.

You know how many people you need to hire, but not which numbers to watch along the way. That is probably what brought you to this page. You post a job, wait to see what happens, and just before the deadline you realize there are not enough applications, or a candidate has declined your offer. You want to manage hiring with numbers so that does not happen.

I have worked as a recruiter for more than 14 years. At a recruitment agency in Japan I placed IT engineers, and at Google I was an in-house recruiter covering Europe, the Middle East and Africa. Today I provide recruitment consulting for SMBs, and I help companies plan their hiring in free recruitment consulting sessions.

  1. First, the difference between a KGI and KPIs, and how a KPI tree works back from the number of hires.
  2. Then the benefits of recruiting KPIs and the 10 metrics, each with its formula, benchmarks (poor, standard, excellent) and an example of how to fix it.
  3. Next, how the KPIs change for mid-career, new graduate and engineer hiring.
  4. After that, how to set your KPIs in 5 steps, and how to track and review them.
  5. The guide ends with the pitfalls: what happens when a number becomes the target.

How it works

How recruiting KPIs work: KGI vs KPI and the KPI tree

You may think a hiring target is enough on its own. But the number of hires does not move until hiring is over. The only numbers you can act on along the way are the ones before it.

KGI vs KPI: hires are the result, KPIs are what you can move along the way

The KGI (key goal indicator) is your end goal, for example "hire 2 salespeople by the end of March." Recruiting KPIs (key performance indicators) are the numbers to hit on the way there, for example 64 applications, 12 interviews and 3 offers.

The difference is when you can move them. You only know the number of hires at the end of hiring, but you can raise applications and interviews with what you do this week. Management frameworks work the same way: each result is paired with the numbers that drive it (Kaplan and Norton, 1996). So "our KPI this quarter is 10 hires" only restates the KGI: it does not tell you what to increase tomorrow.

Recruiting KPI tree: divide by pass rates to find the applications you need

A recruiting KPI tree starts from the number of hires and divides by the pass rate of each stage to get the number needed at the stage before. Hires lead back to offers, offers to interviews, and interviews to applications, so you can draw it like the branches of a tree.

For example, to make 2 mid-career hires at average pass rates, you need 64 applications (doda's job-seeker data). The tree below shows each step.

Recruiting KPI tree for 2 mid-career hires

Worked back from doda's screening and interview pass rates and Mynavi's offer acceptance rate. The round tags show the share that reaches each stage.

  1. 64Applications
  2. 12Interviews18.5%interview
  3. 3Offers20%offer
  4. 2Hires85%accept

Source: calculated from doda, Average Number of Applications of Successful Job Changers, and Mynavi Mid-Career Hiring Survey 2026

This calculation is the basis for turning a hiring plan into KPIs. How to set it up with your own numbers is in how to set recruiting KPIs.

Benefits

Benefits of recruiting KPIs: spot the weak stage before the deadline

The benefit of recruiting KPIs is that you see which stage is short of candidates before the result is in. If you only watch the number of hires, you learn you are short after the deadline. With a KPI tree, the application count in week 3 already tells you that you will miss the target at this pace.

The stage also tells you what to do. Too few applications: the job posting or the hiring channels. Too few move on to interviews: the requirements in the job posting. Offers declined: the terms or the speed of the process. You narrow the fix down to one place. Without numbers, the choice tends to come down to more job ads or more recruitment agencies.

It also makes it easier to ask for help inside the company. You can talk to interviewers in numbers ("4 interviews this month") and to the CEO ("if applications drop below 5 a week, we add a job ad"). Making hiring more efficient also only starts once you know which stage takes your time.

And you can plan your next hire from your own results. In Mynavi's survey, 37.2% of companies look at stage-to-stage pass rates when they review new graduate hiring, but only 12.5% of companies with fewer than 50 employees do (2027 graduates offer survey). In other words, most small companies start the next hire without reviewing the numbers from the last one.

KPI list

List of recruiting KPIs: 10 metrics with formulas and benchmarks

Here are 10 examples of recruiting KPIs. They follow the stages from application to after joining, with each formula, the mid-career average, and the poor and excellent benchmarks.

The hiring process and its KPIs

The KPIs to watch at each stage

  1. Attract

    Ads and scouts

  2. Apply

    Screening

  3. Interview

    First to final

  4. Offer

    Until the answer

  5. Join

    Start date

  6. Retain

    First 6 months

How the poor and excellent levels are set:

  • No-show rate, offer acceptance, fill rate, days to hire: worse than every industry average is poor, better than every industry average is excellent.
  • Scout reply rate: the range of averages on the 5 platforms with the most sends.
  • Cost per hire: the range of averages of the main paid channels.
  • Applications per hire, interview rate, offer rate, early turnover: approximate, set around the average, because no published data shows the spread.
List of recruiting KPIs (formula, average, poor and excellent benchmarks)
MetricFormulaPoorAverageExcellent
KPI 1Applications per hireApplications ÷ hires> 60About 32< 15
KPI 2Scout reply rateReplies ÷ scouts sent< 2.2%6.7%> 16.9%
KPI 3Interview rateInterviews set ÷ applications< 10%18.5%> 30%
KPI 4Interview no‑show rateNo-shows ÷ interviews set> 8.2%5.4%< 3.5%
KPI 5Offer rateOffers ÷ interviews set< 10%20%> 30%
KPI 6Offer acceptance rateAcceptances ÷ offers< 75.7%85.0%> 90.7%
KPI 7Early turnover rateLeavers within 6 months ÷ hires> 15%8.3%< 5%
KPI 8Fill rateHires ÷ hiring target< 60.3%73.3%> 80.5%
KPI 9Days to hireAcceptance date − application date> 36.3 days31.5 days< 26.6 days
KPI 10Cost per hireHiring costs ÷ hires> ¥914,000¥851,000< ¥285,000

Averages are for mid-career hiring (cost per hire: agency average). Each KPI's section names its sources.

The 3 KPIs that matter most

You do not need to track all 10 metrics with the same weight. For an SMB, these are the three to watch first.

  • 1. Applications

    The one number you can raise with this week's work. Everything else follows from it.

  • 2. Offer acceptance rate

    One decline sends you back through about 27 applications of selection.

  • 3. Early turnover rate

    The check that hires made to hit the numbers are still with you.

Applications come first because they are almost the only number the person in charge can raise with this week's work.

Second, the offer acceptance rate is the most expensive stage. By doda's pass rates, each offer sits on top of about 27 applications, and a decline means running that selection again. Declines jumped to 15.0% in 2025 from 9.3% the year before, and reached 18.4% at companies with 3–50 employees.

Third, the early turnover rate checks that chasing numbers has not lowered your hiring bar.

Companies with 50 or fewer employees should also watch a fourth: the interview no-show rate. At companies with 3–50 employees it is 14.3%, about 2.6 times the overall 5.4%.

KPI type 1: Applications

Recruiting KPIs for applications: applications and the scout reply rate

Applications can seem like something you just wait for, but the stage that brings them in has KPIs too. When people apply to your job posting, the top of the tree is applications; when you contact candidates first, it is the scout reply rate.

KPI 1: ApplicationsThe earliest count, tracked every week

This KPI counts the applications received in a period, with the same person counted once. Compare each week's count with the target you worked back from your hires.

Cheat sheet · KPI 1

Applications per hire

What it is

How many applications it takes to make one hire.

Why it matters

It is almost the only number you can raise this week, and every later stage depends on it.

Formula
Applications per hireequalsApplicationsdivided byHires

Applications per hire: benchmarks

Mid-career hiring

Lower is better

Approximate benchmarks around the average. Average calculated from doda, Average Number of Applications of Successful Job Changers

According to doda's data, job seekers applied to 31.9 openings on average before getting an offer. Its screening and first-interview pass rates put one offer at about 27 applications. With most offers accepted, that means about 32 applications per hire. Applicant tracking data from outside Japan puts it at 291 applications per hire (Ashby, 2026).

Estimate

How to fix applications

You need 5 a week but get 2.5

  1. Problem

    2.5 a week

    The plan is 2 hires in 12 weeks and needs 64 applications, but the pace will bring only 30.

  2. Action

    Compare pay and terms with the market, and put the job and the salary in the first 3 lines of the job posting. Add one hiring channel.

  3. Result

    5 a week

    Back above 5 applications a week, the hires you can expect in 12 weeks rise from about 1 to about 2 (at average pass rates).

When applications are short, first check that your pay and terms hold up in the hiring market. How to write the job posting is in the job posting guide.

KPI 2: Scout reply rateThe share of scouts that get a reply

The scout reply rate is the share of the scout messages you send that get a reply. In direct recruiting, where you reach out instead of waiting for applications, it takes the place of applications at the top of the tree.

Cheat sheet · KPI 2

Scout reply rate

What it is

The share of scout messages that get a reply.

Why it matters

When you contact candidates first, it replaces applications at the top of the tree.

Formula
Scout reply rateequalsRepliesdivided byScouts senttimes100

Scout reply rate: benchmarks

Scouting firm send results (2025)

Higher is better

Benchmarks set by the range of averages on the 5 platforms with the most sends (September 2025). Sources: Direct Sourcing, Nikkei's job board

The reply rate changes tenfold with how you send. On Nikkei's job board, only 0.7% of scouts got a reply in September 2026 (direct recruiting guide).

In the send results of Direct Sourcing, a scouting firm (2025), the average was 6.7%, and 8.1% for engineering roles. The same firm's data shows 16.2% in 2018: reply rates fall year after year. Across the 5 platforms with the most sends, averages ranged from 2.2% on Green to 16.9% on YouTrust (September 2025).

Estimate

How to fix the scout reply rate

The same message sent to everyone

  1. Problem

    1.5%

    150 scouts sent, 2 or 3 replies. The messages are not reaching people who fit.

  2. Action

    Stop mass sends. Write each message one by one, with what caught your interest in the person's background in the first 2 lines. Send only to people who meet the requirements.

  3. Result

    6.7%

    At the scouting firm average, the same 150 scouts bring about 10 replies.

KPI type 2: Selection

Recruiting KPIs for selection: interview, no-show and offer rates

Applications do not guarantee hires. At the selection stage, watch three rates: how many move on to an interview, how many show up, and how many get an offer. Of these, the one that breaks down most at SMBs is how many show up.

KPI 3: Interview rateThe share of applications that reach an interview

The interview rate is the share of applicants you invite to an interview. It means almost the same as the document screening pass rate, and shows whether your job posting attracts people who fit.

Cheat sheet · KPI 3

Interview rate

What it is

The share of applications that lead to an interview.

Why it matters

It shows whether your job posting attracts people who fit.

Formula
Interview rateequalsInterviews setdivided byApplicationstimes100

Interview rate: benchmarks

Mid-career hiring

Higher is better

Approximate benchmarks around the average. Average: doda, Average Number of Applications of Successful Job Changers

By doda's data pass rates, about 1 application in 5 (18.5%) leads to an interview. Mynavi's survey of job changers finds a higher 37.3%, but it also counts openings introduced by an agency as applications (Mynavi).

Estimate

How to fix the interview rate

Many applicants miss the requirements

  1. Problem

    8%

    Of 50 applications, 4 reached an interview. Many applicants lack the required experience.

  2. Action

    Put the must-have requirements and the salary at the top of the job posting so people who do not fit do not apply.

  3. Result

    18.5%

    At the average rate, the applications needed for 2 hires fall from about 147 to about 64.

KPI 4: Interview no-show rateThe share who skip an interview without notice

The interview no-show rate is the share of scheduled interviews the candidate misses without telling you. It is especially high at SMBs and quietly cuts the number of interviews you hold.

Cheat sheet · KPI 4

Interview no-show rate

What it is

The share of scheduled interviews missed without notice.

Why it matters

It is highest at small companies, and quietly cuts the interviews you hold.

Formula
Interview no-show rateequalsNo-showsdivided byInterviews settimes100

Interview no-show rate: benchmarks

Mid-career hiring (2025 results)

Lower is better

Benchmarks set by the range of industry averages (30+ companies each). Source: Mynavi Mid-Career Hiring Survey 2026

In Mynavi's survey, 5.4% of interviews in 2025 were no-shows. At companies with 3–50 employees, though, the rate was 14.3%, about 2.6 times the overall figure. By industry the averages all sit between 3.5% and 8.2%, so small companies do worse than every industry average. Companies that hold interviews online only had just 1.6% (59 companies answered).

Estimate

How to fix the no-show rate

1 in 7 candidates does not come

  1. Problem

    14.3%

    7 interviews scheduled, and 1 candidate does not show up. Times are arranged by email a few days after the application.

  2. Action

    Set the interview time on the day of the application and send a reminder the day before. Let candidates take the first interview online.

  3. Result

    5.4%

    Down to the average, no-shows fall from about 3 to about 1 for every 20 interviews.

KPI 5: Offer rateThe share of interviewed candidates who get an offer

The offer rate is the share of candidates invited to an interview who receive an offer. It reflects both how strict your interviews are and the quality of the people who come.

Cheat sheet · KPI 5

Offer rate

What it is

The share of interviewed candidates who get an offer.

Why it matters

It tells you whether your interview bar is too strict or too loose.

Formula
Offer rateequalsOffers madedivided byInterviews settimes100

Offer rate: benchmarks

Mid-career hiring

Higher is better

Approximate benchmarks around the average. Average: doda, Average Number of Applications of Successful Job Changers

By doda's data pass rates, about 1 interview in 5 (20%) ends in an offer. A higher offer rate is not always better, though: an offer rate raised by lowering the bar comes back as early turnover (pitfalls).

Estimate

How to fix the offer rate

The result depends on the interviewer

  1. Problem

    8%

    2 offers from 25 interviews. The same candidate gets opposite ratings from different interviewers.

  2. Action

    Set questions and rating criteria from what the job needs, ask every candidate the same questions, and decide against the criteria.

  3. Result

    20%

    At the average rate, the interviews needed for 2 hires fall from about 29 to about 12.

How to set questions and rating criteria is in the interview questions guide.

KPI type 3: Offers and joining

Recruiting KPIs for offers and joining: offer acceptance and early turnover

Hiring is not over once you make an offer. It only works when the candidate accepts and then stays after joining. Both numbers tend to be worse at smaller companies.

KPI 6: Offer acceptance rateThe share of offers accepted

The offer acceptance rate is the share of candidates who accept your offer, or 100% minus the offer decline rate. A decline sends you back through selection, which makes this the most expensive stage.

Cheat sheet · KPI 6

Offer acceptance rate

What it is

The share of offers that candidates accept.

Why it matters

One decline sends you back through about 27 applications of selection.

Formula
Offer acceptance rateequalsOffers accepteddivided byOffers madetimes100

Offer acceptance rate: benchmarks

Mid-career hiring (2025 results)

Higher is better

Benchmarks set by the range of offer decline rates by industry (30+ companies each). Source: Mynavi Mid-Career Hiring Survey 2026

In Mynavi's survey, 15.0% of mid-career offers were declined in 2025 (an 85.0% acceptance rate), up from 9.3% the year before. Companies with 3–50 employees saw 18.4% and those with 51–300 saw 18.9%: SMBs get more declines.

By industry it ranges from 9.3% in distribution, retail and food service to 24.3% in finance, insurance and consulting. Data from an applicant tracking system outside Japan also puts the average acceptance rate at 78%, and 73% for technical roles (Ashby, 2024).

Declines seem to be about the job itself more than the terms. In a new graduate survey, the top reason companies gave for students declining was "the work or placement they wanted" (20.7%), followed by "the job type they wanted" (18.3%) and "salary" (12.9%) (Mynavi, 2027 graduates).

Estimate

How to fix offer acceptance

5 offers, only 3 accepted

  1. Problem

    60%

    5 offers made, 3 accepted. The reasons: "the work was not what I imagined" and "I chose another company."

  2. Action

    Before the offer, explain the team's actual work and the salary in detail so nothing comes as a surprise. Give one week to answer, and set up a talk with team members during that week.

  3. Result

    85%

    At the average rate, the offers needed for 2 hires fall from about 3.3 to about 2.4, and applications from about 90 to about 64.

KPI 7: Early turnover rateThe share of hires who leave within 6 months

The early turnover rate is the share of hires who leave within a set period after joining. The figures recruitment agencies publish also count leavers within 6 months. When count-based KPIs push you to hire in a rush, the side effect shows up here.

Cheat sheet · KPI 7

Early turnover rate

What it is

The share of hires who leave within 6 months of joining.

Why it matters

It checks that chasing numbers has not lowered your hiring bar.

Formula
Early turnover rateequalsEarly leaversdivided byHirestimes100

Leavers within 6 months: benchmarks

People placed in permanent jobs by agencies (fiscal 2024)

Lower is better

Approximate benchmarks around the average. Average: figures published by recruitment agencies on the Ministry of Health, Labour and Welfare portal, all agencies added up

For mid-career hires, the figures recruitment agencies publish on the Ministry of Health, Labour and Welfare's recruitment services portal are a usable benchmark. Adding up all agencies' figures, 8.3% of the people placed in permanent jobs through agencies in fiscal 2024 left within 6 months. For another 5.4%, the agency does not know whether they left, so the real rate may be a little higher.

Among new graduates, 31.2% of university graduates who started work after graduating in March 2023 left within 3 years (Ministry of Health, Labour and Welfare). By workplace size, the smaller the workplace, the higher the rate.

Turnover within 3 years of university graduates, by workplace size

March 2023 graduates

  1. Under 5 staff58.2%
  2. 5–29 staff50.6%
  3. 30–99 staff39.7%
  4. 100–499 staff31.7%
  5. 500–999 staff29.0%
  6. 1,000+ staff24.2%

Source: Ministry of Health, Labour and Welfare, turnover of new graduates (March 2023 graduates)

Small workplaces lose the most. At workplaces with 5–29 staff it is 50.6%, about twice the 24.2% at the largest workplaces (1,000 staff or more), and higher even than the highest industry, accommodation and food services (49.9%).

Estimate

How to fix early turnover

4 hires, 2 left within 6 months

  1. Problem

    50%

    4 people hired in a year; 2 left within 6 months, saying "the job was not what I was told."

  2. Action

    In interviews, describe the hard parts of the job concretely too, and explain them again when you make the offer.

  3. Result

    25%

    With 1 leaver instead of 2, there is 1 fewer person to replace next time: about 32 applications of selection you no longer need.

Researchers call this up-front honesty about the hard parts a "realistic job preview" (RJP). Pooling 52 studies, one meta-analysis measured a small but real drop in turnover, and no fall in offer acceptance. Of the ways to deliver it, telling candidates face to face in the interview had the strongest effect (Earnest et al., 2011).

KPI type 4: Planning

Recruiting KPIs for planning: fill rate, days to hire and cost

The last three measure the result of hiring as a whole and its conditions. You do not track them weekly, but record them as the report card at the end of each hiring round.

KPI 8: Fill rateThe share of your hiring target you actually hired

The fill rate is the share of your hiring target you actually hired. It shows directly how far you reached your KGI.

Cheat sheet · KPI 8

Fill rate

What it is

The share of your hiring target you actually hired.

Why it matters

It shows directly how far you reached your hiring goal.

Formula
Fill rateequalsHiresdivided byHiring targettimes100

Fill rate: benchmarks

Mid-career hiring (2025 results)

Higher is better

Benchmarks set by the range of industry averages (30+ companies each). Source: Mynavi Mid-Career Hiring Survey 2026

In Mynavi's survey, mid-career hires reached 73.3% of the hiring target across the surveyed companies in 2025. Companies with 3–50 employees reached 58.7%, below every industry average (60.3–80.5%). In new graduate hiring, the fill rate for 2027 graduates (offers ÷ openings) was 78.3% (Mynavi).

Estimate

How to fix the fill rate

2 new graduate offers for a target of 2

  1. Problem

    50%

    The goal was 2 new graduates. 2 offers went out and 1 was declined.

  2. Action

    Plan for the 36.6% decline rate on informal offers (nainaitei) that companies report, and make more offers from the start.

  3. Result

    100%

    2 ÷ 0.634 is about 3.2. With a plan of 4 offers, you can expect about 2.5 hires, enough for the target of 2.

KPI 9: Days to hireFrom application to offer acceptance

Days to hire is the number of days from application to offer acceptance. The longer it takes, the more often another company gets there first and candidates decline. The definition and targets for each stage are in the time to hire guide.

Cheat sheet · KPI 9

Days to hire

What it is

The number of days from application to offer acceptance.

Why it matters

The longer it takes, the more candidates another company wins first.

Formula
Days to hireequalsAcceptance dateminusapplication date

Days from application to offer acceptance: benchmarks

Mid-career hiring, companies that interview online (2025 results)

Lower is better

Benchmarks set by the range of industry averages (30+ companies each). Source: Mynavi Mid-Career Hiring Survey 2026

In Mynavi's survey, it took an average of 31.5 days from application to offer acceptance at companies that interview online. 23.2% of companies finish in under 2 weeks, but 33.4% take a month or more.

Estimate

How to fix days to hire

2 weeks between interviews

  1. Problem

    45 days

    45 days from application to acceptance. Two weeks pass between the 2 interviews, and the result takes a week.

  2. Action

    Offer interview times on the day of the application, keep interviews within a week of each other, and give results within 2 business days.

  3. Result

    About 31 days

    Cutting about a week from both the gap between interviews and the wait for results brings 45 days down to about 31, the average.

KPI 10: Cost per hireThe hiring cost for each person

Cost per hire is the total you spent on hiring, split across the people you hired. It changes tenfold with salary and hiring channel, so compare it by hiring channel. The benchmark uses the averages of the main paid channels.

Cheat sheet · KPI 10

Cost per hire

What it is

Your total hiring cost divided by the number of hires.

Why it matters

Compared by channel, it shows where your money works hardest.

Formula
Cost per hireequalsTotal hiring costsdivided byHires

Cost per hire: benchmarks

Per full-time hire (2021)

Lower is better

Benchmarks set by the range of averages of the main paid channels. Source: Ministry of Health, Labour and Welfare commissioned survey (2021, 1,142 employers)

In a 2021 survey commissioned by the Ministry of Health, Labour and Welfare, the average cost of one full-time hire was ¥285,000 through job sites, ¥851,000 through recruitment agencies and ¥914,000 through scouting. Costs by channel, the formula and a calculator are in the cost per hire guide.

Estimate

How to fix cost per hire

Both hires through a recruitment agency

  1. Problem

    ¥1.75M

    2 hires at a ¥5 million salary through an agency charging 35%: ¥1.75 million per hire.

  2. Action

    Add job sites and scouting, and keep the agency for the one hire that is hard to fill.

  3. Result

    About ¥1.02M

    If 1 hire comes through a job site at the ¥285,000 average, the average for the 2 falls to about ¥1.02 million.

Hiring types

Recruiting KPIs by type of hiring: new grads need about twice the applications

Even for the same 2 hires, the shape of the KPI tree changes with mid-career, new graduate or engineer hiring. Compare the applications you need for one hire, and the difference is clear.

Applications needed for 1 hire

Calculated from survey averages

  1. Mid-career (applications)About 32
  2. New graduates (entries)About 70

Source: calculated from doda, Mynavi Mid-Career Hiring Survey 2026 and Mynavi New Graduate Offer Survey (2027 graduates)

In mid-career hiring, the KPIs center on applications and the offer acceptance rate. Applications arrive a few at a time all year, so track them as a weekly count and review pass rates after each hiring round.

In new graduate hiring, the KPIs are entries and the informal offer (nainaitei) decline rate. In Mynavi's 2027 graduates survey, 36.6% of the informal offers companies made were declined (34.3% at companies with fewer than 50 employees), and each offer took 44.1 entries and 5.5 first interviews.

Allowing for declines, one new joiner takes about 70 entries. The hiring season is fixed and you cannot make up for it later, so make extra offers to allow for declines (fill rate example).

In engineer hiring, waiting for applications rarely works. doda's job openings-to-applicants ratio was 2.89 overall in September 2026 but 12.04 for engineers (IT and telecom): about 12 job openings compete for each engineer who wants to change jobs. So the top of the tree becomes scouting, and the KPIs are scouts sent and the reply rate.

Steps

How to set recruiting KPIs: work back in 5 steps

How do you set KPIs? Start from the number of hires and work back through the selection stages one by one, in 5 steps.

  1. Set the KGI

    Decide how many people, of what profile, to hire by when. Take the deadline from the date you need them to start, and the number from your vacancies and business plan. SMART, the well-known guide to writing goals, started as 5 points: specific, measurable, assignable, realistic and time-related (Doran, 1981).

  2. Define the stages and how to count them

    List the stages of your selection process (application, interview, offer, acceptance and so on) and decide what counts as one at each. Settle here whether scout replies count as applications, and whether interviews count when set or when held.

  3. Set a pass rate for each stage

    If you have results from your last hire, use them. If not, start from the averages in the KPI list. Split by hiring channel (recruitment agencies, job ads, employee referrals and so on) if you can.

  4. Work back to weekly numbers

    Divide the number of hires by the offer acceptance rate, the offer rate and the interview rate, in that order, to get the applications you need. Divide by the weeks until the deadline for your weekly target. For 2 hires in 12 weeks, that is 64 applications, or a little over 5 a week.

  5. Pick one KPI to track, and pair it with a retention metric

    Track one number each week that the person in charge can raise on their own: usually applications or scouts sent. Pair it with one metric that checks retention after joining (pitfalls).

Split by hiring channel because pass rates differ widely between channels. Data from an applicant tracking system outside Japan shows that 52% of candidates who came through employee referrals passed the first screen, well above the 35% overall (Ashby, 2026).

A study of referrals also found that referred hires accept offers more often and leave less (Burks et al., 2015). If you mix channels, your pass rate rises in months with many referrals, even when nothing else has changed.

Tracking and review

Tracking and reviewing recruiting KPIs: counts weekly, pass rates each round

Even with recruiting KPIs set, the numbers do not help if you record them wrongly or look at them at the wrong times. The key is to look at counts and pass rates at different intervals.

How to build the tracker: one row per candidate, totals by week

A recruiting KPI tracker has two parts: a record for each candidate and a weekly summary. The candidate record keeps one row per person with the application date, the hiring channel, the result and date at each stage, and the reason if they withdrew. The weekly summary lines up that week's applications, interviews, offers and acceptances. With an ATS (applicant tracking system), both are counted for you.

With small numbers, do not judge by pass rates

Look at counts every week, and pass rates only after a hiring round ends. SMB hiring involves small numbers, and rates swing widely by chance.

  • 30% apparent pass rate

    3 of 10 passed

    11–60%Range of the true pass rate

  • 30% apparent pass rate

    9 of 30 passed

    17–48%Range of the true pass rate

  • 30% apparent pass rate

    15 of 50 passed

    19–44%Range of the true pass rate

  • 30% apparent pass rate

    30 of 100 passed

    22–40%Range of the true pass rate

95% confidence interval (Wilson method).

If you interview 10 people and make offers to 3, all you can say statistically is that the true offer rate lies somewhere between about 10% and 60%. If the next 10 bring 2 offers or 5, you still cannot say anything in your process changed. So look at counts each week, and compare pass rates with the last round only once a full hiring round (a few months to a year for mid-career) has built up.

When you compare, match the role, the hiring channel and the length of the period with the last round. A pass rate that mixes engineers and salespeople moves just because the mix of applications changes.

Review the KGI too, after each hiring round

In the review, first find the stage that is most short of numbers right now, and move the KPI you track weekly in the next round to that stage. If applications were enough, track the offer acceptance rate next, for example.

If fixing the stages still does not get you there, review the KGI itself. Many companies cannot hire the number they need. In the Recruit Works Institute's survey, 47.0% of companies "secured" the mid-career hires they needed in the second half of fiscal 2025, fewer than the 51.5% that "could not."

Decide in your review of hiring results whether to extend the deadline, hire fewer people, or relax the requirements and train after joining.

Pitfalls

Recruiting KPI pitfalls: when a number becomes the target, quality drops

The weakness of KPIs is that the number easily becomes the goal itself. "When a measure becomes a target, it ceases to be a good measure" is the anthropologist Strathern's rewording of a point made by the economist Goodhart (University of Cambridge explainer).

In hiring, make offers the target and people lower the bar to make offers; make interview pass rates the target and interviewers rate more leniently. The numbers are hit, but the hiring fails.

The way to prevent this is to always pair count-based KPIs with a retention metric after joining. This matters most for SMBs: as the early turnover rate section showed, smaller workplaces lose more of their new hires.

To measure quality of hire, the usual approach combines performance after joining, retention and manager satisfaction; in a survey of recruiters outside Japan, 66%, 60% and 44% used each (LinkedIn, 2025). Yet in the US, only 20% of organizations measure it (SHRM, 2025). For an SMB, two numbers are enough: the share of hires who pass probation, and the manager's rating at 3 months.

When you make recruiting KPIs someone's personal goals, use only numbers they can raise with their own actions, such as scouts sent or days from application to first contact. The number of hires and the offer acceptance rate depend heavily on salary and on what other companies do, and using them in personal reviews pushes people to pick only the candidates who are easy to get through.

FAQ

Frequently asked questions about recruiting KPIs

How many recruiting KPIs should you set?

Track one KPI each week, and watch as many numbers as you have selection stages. Make the one you track the count at the stage that is shortest right now, and just check the rest for changes. Some experts compare KPIs to a traffic light and say to narrow them to one (Ryuichiro Nakao). Management research has long recommended narrowing measures down to the few that truly matter (Kaplan and Norton, 1992).

What is the difference between recruiting KPIs and OKRs?

Recruiting KPIs are numbers worked back from one result, the number of hires, while OKRs are a goal-setting method that pairs an objective with its key results. If you only manage hiring, a KPI tree is enough. If your company uses OKRs, make the number of hires one of the key results and put the recruiting KPIs under it.

What is the average offer decline rate in Japan?

It depends a lot on how you count. The share of a company's offers that were declined is 15.0% in mid-career hiring (2025) and 36.6% in new graduate hiring (2027 graduates), both from Mynavi. The "64.5%" often quoted for new graduates is the share of students who declined at least one offer, not a company's decline rate (job search process survey, 2026 graduates). Compare your KPIs with the company-side figures.

Do you need recruiting KPIs if you hire only 1 person a year?

Yes, but two numbers are enough: applications and retention. With 1 hire a year, each stage only sees a few candidates, and you cannot tell differences in pass rates from chance. Watch applications every week, check that the hire passes probation, and look at pass rates only after several hires have built up (with small numbers).

Is there a recruiting KPI template or Excel sheet?

A spreadsheet with two parts is enough: one row per candidate, and a weekly summary. Which columns to include is in how to build the tracker.

DIY Hiring Kit

Everything You Need To Hire On Your Own

A downloadable hiring framework for small businesses in Japan, with step-by-step guides, spreadsheet ATS, LinkedIn scouting system, interview questions bank, offer letter and employment contracts, and a ton more!

1Dashboard11Checklists8Spreadsheets30Template docs4Weeks of Q&ABilingual

  • Download every file as Word, Google Docs, Excel or Google Sheets

About the author

Emmanuel Gendre, Founder of SaiyouTeam

Emmanuel Gendre

Founder, SaiyouTeam · HR and recruiting advisor to SMBs in Japan

  • Former Google recruiter
  • 14 years in talent acquisition

Emmanuel advises SMBs on HR and recruiting in Japan. Before founding SaiyouTeam, he spent 14 years in talent acquisition, including as a recruiter at Google. This guide is part of that advisory work: it brings together how to set recruiting KPIs, their benchmarks, and how to track and review them.