How Many Candidates Are Interviewed per Hire? 2025 and 2026 Benchmarks by Role and Stage

By Sparsh Goyal, Founder at The Cognitive. Published 2026-10-04 · Last updated 2026-10-05

• The average number of candidates interviewed per hire is 11.7 for business roles and 17.6 for technical roles in Ashby's 2026 data, about 4 at small businesses (CareerPlug), and 20 or more if you count interview sessions instead of people (Gem).
• The benchmarks disagree mainly because they count different things: people or sessions, with or without the recruiter screen, at tech companies or small businesses.
• Measure your own as unique candidates interviewed divided by hires for one cohort of roles, then compare by role family; I run The Cognitive, whose live AI first round takes the first conversation off your team's calendar.

The average number of candidates interviewed per hire is about 12 for business roles and about 18 for technical roles, according to Ashby's 2026 data on 247,000 jobs. Small businesses interview far fewer: CareerPlug's 27% interview-to-hire rate works out to about 4 people per hire. Gem counts interview sessions instead of people and gets 20 per hire.

The spread is real, and most of it comes from what each source counts. This page puts the numbers side by side, splits them by role and by stage, and shows how to calculate your own. The Cognitive runs the first round as a live AI interview, which changes how many people you can afford to interview at the top. I cover that near the end, with facts only.

Full disclosure: I run The Cognitive. Every benchmark here comes from the publisher's own report, read on 4 October 2026. This page is updated annually.

What is the average number of candidates interviewed per hire?

For a company hiring like Ashby's customers, plan on 11.7 candidates interviewed for each business hire and 17.6 for each technical hire. Ashby's 2026 Recruiter Productivity report calls the section "Teams are interviewing 52% more applicants", and its chart counts applications interviewed per hire. Both numbers are up sharply from 2021, when they stood near 8 and 11.

Here is every published figure I could verify, with what each one counts:

So the honest answer to "how many candidates are interviewed per hire" is a range: about 4 at small businesses, about 12 to 18 at growth and enterprise companies, and 20 to 36 if you count every interview session instead of every person.

Why do the benchmarks disagree?

4 things split the numbers, and you need to know which one applies before you compare yourself to any of them.

People or sessions

Ashby counts candidates who were interviewed. Gem counts interviews. A finalist who meets 5 people across 3 rounds adds 1 to Ashby's count and several to Gem's. That alone explains most of the gap between 17.6 and 35.

Who is in the sample

Ashby's and Gem's figures come from companies that run their hiring on Ashby or Gem, which in my experience skews toward tech and growth companies with structured loops. CareerPlug's data comes from more than 60,000 small business owners, many hiring hourly staff after one conversation. A restaurant and a Series C software company should not share a benchmark.

Whether the recruiter screen counts

Some teams log the 20-minute recruiter call as an interview. Others start counting at the hiring manager round. Ashby reports recruiter screens as a stage with a pass rate of about 35%, so I read its interview counts as including them. Ashby doesn't spell this out. If your ATS doesn't log screens, your ratio will look better than Ashby's for no real reason.

A check that the numbers hang together

Ashby also reports that 10.4% of interviewed business candidates and 7.3% of technical ones reach an offer. Gem reports that 82% of offers are accepted. Multiply them and you get about 11.7 business candidates and 16.7 technical candidates interviewed per hire. That's my arithmetic across 2 different samples, but it lands close to Ashby's direct counts, which gives me confidence in the 12 and 18 range.

How has the number changed since 2021?

Every source agrees on the direction: teams interview more people for each hire than they did 5 years ago. The figures for 2025 and early 2026 are the highest in each series.

• Ashby: about 8 business candidates interviewed per hire in 2021, 11.7 by early 2026, up 36%. Technical went from about 11 to 17.6, up 52%.
• Gem: 14 interviews per hire in 2021, 20 by the end of 2024, up 42%. Its 2026 report, covering data to May 2025, puts the rise since 2021 at 33%.
• Applications: Ashby's applications per hire went from about 100 in early 2021 to 291, and stayed above 300 for all of 2025.

Here's the part most summaries miss. A single applicant's odds of an interview fell from 7 to 8% in 2021 to between 3.6% and 4.7% now, according to Ashby. Yet once someone is in the process, their odds of an offer have risen above 2021 levels. Teams are screening harder at the top and interviewing more of the people who pass.

The hired candidate barely noticed. Ashby found the time a hired person spends in interviews has stayed flat since 2021, at about 2.4 hours for business roles and 3.5 for technical. In Ashby's words, the extra burden "has fallen on recruiting teams, not on the people moving through the funnel." That burden is what interviewer hours per hire measures.

Gem ties the extra interviews to slower hiring: its average time to hire rose 24%, from 33 to 41 days, over the same years.

How many candidates are interviewed per hire by role type?

Role type matters more than company size. Ashby split its data by department for Q1 2025 to Q1 2026:

The gap between Data and Customer Support is more than 10 people per hire and almost 16 interview hours. If one recruiter covers both, a single capacity number will be wrong for both.

Technical roles are also less predictable. About 23% of technical hires in Ashby's data took 5 or more hours of total interview time, against 6% of business hires. Gem's figure of 35 to 36 interviews per technical hire points the same way.

Hourly and small-business roles sit at the low end. CareerPlug's 27% interview-to-hire rate means about 1 hire for every 3 or 4 people interviewed. Its applicants per hire vary by industry, from about 234 in automotive to 57 in education and childcare.

Executive roles are a thin slice in every dataset. Ashby found only about 35% of executive hires come from inbound applicants, so most executive interviews start with people someone went out and found. SHRM puts median executive time to fill at about 45 days in 2025, down from 60 in 2022.

How many candidates reach each interview stage?

No public report gives a full stage-by-stage count per hire, so I built one from the stage pass rates that are published. Treat it as a planning sketch for a business role.

• Recruiter screen: about 11.7 candidates per hire. Ashby says about 35% pass this stage.
• First round with the hiring manager: about 4 candidates per hire, which is 11.7 times 35%. This assumes every interviewed candidate starts with a recruiter screen.
• Onsite or final loop: fewer than 4. Ashby doesn't publish an onsite pass rate as a number in its text, so I won't invent one.
• Post-onsite to offer: Ashby reports 95% pass-through at the post-onsite stage and 81% at offer.
• Offer to hire: about 1.2 offers per hire, because 82% of offers are accepted (Gem).

Gem's top-of-funnel view says the same thing from the other end: only 8% of applicants advance past initial screening and 0.5% receive an offer. That's roughly 1 offer for every 16 people who got past the screen.

For the hired person, the process looks much smaller than these totals. Ashby counts 4 interview events per business hire and 5 per technical hire, adding up to about 2.4 and 3.5 hours for the candidate. Technical candidates take about 18 days from first to last interview, business candidates about 14. Our recruitment funnel conversion rates guide covers the percentage at every stage; this page sticks to counts and hours.

How do you calculate candidates interviewed per hire?

Divide the number of unique candidates who had at least one interview by the number of hires from the same group of roles. Use a cohort, such as every role opened in Q1 and closed by Q3, so the people and the hires belong to the same searches.

Candidates interviewed per hire = unique candidates interviewed ÷ hires

Count each person once, no matter how many rounds they did. Leave out no-shows, and decide up front whether the recruiter screen counts. Write that choice down so next year's number is comparable.

Interview to hire ratio

The interview to hire ratio is the same data turned upside down: hires ÷ candidates interviewed, shown as a percentage. If you interviewed 40 people and hired 4, your interview-to-hire ratio is 10%, and your candidates interviewed per hire is 10. CareerPlug's 27% is the small-business version of this number.

Selection ratio formula

The selection ratio is hires ÷ total applicants. It measures how selective the whole funnel is, not just the interview stage. Gem's figure of roughly 1 hire per 200 applications is a selection ratio of 0.5%, and Ashby's 291 applications per hire works out to about 0.34%.

Interviews per hire and interview hours per hire

Count every scheduled session for the cohort, then divide by hires. Do the same with interviewer hours: sum the length of each session times the number of interviewers in it. Interview hours per hire is the number your hiring managers actually feel, and it's the one that tells you whether you can open another role next quarter.

A worked example

These numbers are made up to show the method. A team closes 6 sales roles in a quarter. 74 unique people had at least one interview, across 151 sessions, with 118 interviewer hours in total.

• Candidates interviewed per hire: 74 ÷ 6 = 12.3
• Interview to hire ratio: 6 ÷ 74 = 8.1%
• Interviews per hire: 151 ÷ 6 = 25.2
• Interviewer hours per hire: 118 ÷ 6 = 19.7

12.3 people per hire is close to Ashby's business benchmark. 19.7 interviewer hours is well above Ashby's 12.2, which points to panels that are too large rather than too many candidates. Our recruiting metrics guide shows how to put these next to time to fill and cost per hire on one dashboard.

What changes the number of candidates interviewed per hire?

Where candidates come from

Source mix moves the ratio more than anything else. In Ashby's referrals study, 32% of interviewed internal candidates reached an offer, against 16% of referred and 8% of agency candidates. That's about 3, 6 and 12 or 13 people interviewed per offer, by my division. Gem found referrals convert at 11 times the rate of inbound applicants. A role filled from referrals needs far fewer interviews than one filled from a job board. Our candidate sourcing channels comparison covers which channel suits which role.

How well the screen works

Application volume keeps rising. Ashby counts 291 applications per hire, against about 100 in early 2021, and Greenhouse says applications per company have grown 129% since 2023. In LinkedIn's survey of 8,035 HR professionals, 73% said fewer than half of the applications they get meet all the listed criteria. When the screen lets weak fits through, the interview count per hire climbs and interviewer hours go with it.

How well interviewers agree

Ashby found that about 38% of scorecard pairs differ by at least a point, and nearly half of those cross the line between 2 and 3 on its 1 to 4 scale. That's the line between yes and no. When interviewers disagree, teams add another round or another candidate to settle it. A written rubric per role cuts that drift. Our free interview rubric generator drafts one from a job description.

Seniority and role type

Technical roles take more interviews, more hours and one more interview event, which Ashby says adds roughly a week. Ashby's operations report also found senior roles take 37% longer to fill than junior ones. Set the benchmark by role family, not company-wide.

Speed

Slow loops lose finalists, and every lost finalist sends you back to interview more people. Ashby's median scheduling confirmation is 3.7 hours with its automated scheduler and 5 hours by hand. SHRM found extra-large organizations spend nearly twice as long conducting interviews as the median organization.

How many interviews is too many?

For the candidate, more than 5 interview events is above what Ashby sees for hired technical candidates, and more than 4 is above the business norm. Each extra round adds about a week. Hired candidates spend about 2.4 hours in interviews for business roles and 3.5 hours for technical roles, and those numbers have stayed flat since 2021.

For the team, look at your candidates interviewed per hire against the role benchmark in the table above. If a sales role needs 25 people per hire when Ashby's business average is 11.7, the problem usually sits upstream: an unclear intake, a weak screen or a hiring manager whose bar moves. Adding interviewers fixes none of those.

How does a live AI first round change interview capacity?

A live AI first round takes the first conversation off your team's calendar, so the number of people you can screen is no longer capped by interviewer hours. Greenhouse said in May 2026 that fewer than 7% of applicants get an interview at all, and bought an AI interviewing company so that more applicants get a conversation at that first stage.

Here is how the first round works in The Cognitive. You import candidates from a stage in your ATS, or source them inside the product, and invite any number of them in one go. Each person books their own slot without creating an account and joins a live, two-way AI video interview of 10 or 20 minutes, in any of 9 languages. The rubric is fixed for the role and the questions adapt live to each answer. You get a 1 to 5 score per criterion, a weighted score out of 100, a suggested verdict, the transcript and the recording. Nothing is rejected automatically. A recruiter decides who goes to round 2.

What changes in your numbers: candidates interviewed per hire can go up at the first round without adding human hours, while interviewer hours per hire fall, because people only meet your team from round 2. I won't give you a ratio for our product. We don't publish one, and yours will depend on the role and the source mix.

The 5-minute recording shows a live AI interview for a go-to-market role. The AI asks the candidate about an outbound email campaign they ran, then follows up on what drove the reply rates and how they qualified the leads that came back.

2 limits to know. The Cognitive is not an ATS, so it doesn't calculate funnel metrics or schedule your human rounds; you still measure the ratio in your ATS. And it writes a note with the score, summary and report link back only for candidates you imported from the ATS. Our AI interviewer page covers the rest. AI Interview starts at $99/month.

Interview more candidates without more interviewer hours Invite a whole ATS stage to a live AI first round and read a scored report for each person. Start free

How do you start tracking candidates interviewed per hire?

Pick last quarter's closed roles and pull 3 numbers per role family: unique candidates interviewed, interview sessions and interviewer hours. Divide each by hires. Put the results next to Ashby's 11.7 and 17.6, and next to CareerPlug's 3.7 if you hire hourly staff.

Then find your biggest gap and work backwards. Too many people per hire points at the screen or the source mix. Too many hours per hire with a normal headcount points at panel size. Fix one, and measure the same cohort again next quarter.

If the first round is where your hours go, try a live AI first round on one open role. The free trial comes with 100 sourcing credits.

Start with one role this week Source or import candidates, run the first round with AI and keep your team's hours for the finalists. Start free

Sources

• Ashby, "Recruiter Productivity | 2026 Talent Trends Report", Ben Perry, 28 April 2026: ashbyhq.com (read 4 October 2026)
• Ashby, "Recruiting Operations Benchmarks | 2026 Talent Trends Report", Joel Westmark, 7 May 2026: ashbyhq.com (read 4 October 2026)
• Ashby, "Are referred candidates more likely to get hired?", 16 May 2025: ashbyhq.com (read 4 October 2026)
• Ashby, "How many hires really come from inbound?", 17 October 2025: ashbyhq.com (read 4 October 2026)
• Gem, "Key takeaways from the 2026 Recruiting Benchmarks Report", SJ Niderost, 1 December 2025: gem.com (read 4 October 2026)
• Gem, "10 takeaways from the 2025 recruiting benchmarks report", SJ Niderost, 16 January 2025: gem.com (read 4 October 2026)
• Gem, 2026 Recruiting Benchmarks report landing page: www-static.gem.com (read 4 October 2026)
• CareerPlug, "Recruiting Metrics and Benchmarks You Should Know", 2025 Recruiting Metrics Report (2024 data): careerplug.com (read 4 October 2026)
• SHRM, "2025 Recruiting Benchmarking" data brief: shrm.org (read 4 October 2026)
• Greenhouse, "Greenhouse has entered into a definitive agreement to acquire Ezra AI Labs", 5 May 2026: greenhouse.com (read 4 October 2026)
• LinkedIn, press release on hiring research, 15 January 2025: news.linkedin.com (read 4 October 2026)

All posts · thecognitive.io