Recruiting Funnel Conversion Rates by Stage: 2026 Benchmarks

Data as of October 2026. Sources: Ashby Talent Trends (Q1 2026), Gem Recruiting Benchmarks 2026 (data June 2021 to May 2025), Appcast 2026 Recruitment Marketing Benchmark Report (2025 data), Employ Hiring Benchmarks (2025), CareerPlug (2024), Gem Email Outreach Benchmarks (2025), Recruiterflow (2026), Gartner (4Q25). Updated October 2026.

What are normal recruiting funnel conversion rates in 2026?

Recruiting funnel conversion rates measure the share of candidates who move from one hiring stage to the next. The table below lists each stage with the most recent published benchmark we could verify.

The short version: very few applicants get past the first screen, but once a candidate is interviewing, the odds improve fast. Most of the loss happens at the top.

Why do the same stages show such different numbers?

The sources count different things. Ashby measures interview to offer across every interviewed application, so a candidate who failed a first phone interview counts. Employ and Appcast report narrower stage to stage moves, so their rates run higher.

Samples differ too. Appcast only tracks applicants who came through paid job ads at about 1,200 US employers. Ashby and Gem skew toward tech and venture backed companies. CareerPlug covers more than 60,000 small businesses. Pick the benchmark whose sample looks most like your own hiring.

How many candidates does it take to make one hire?

The cleanest single source funnel comes from the Appcast 2026 report. It tracks median conversion from job ad click to hire for 2025, across 302 million clicks and 27 million applies.

Working backward from one hire with Appcast’s stage rates gives this Work Insiders model: 1 hire ÷ 0.7521 = 1.33 offers. 1.33 ÷ 0.4045 = 3.29 interviews. 3.29 ÷ 0.6133 = 5.36 screens. 5.36 ÷ 0.1498 = 35.8 applications. 35.8 ÷ 0.0466 = about 768 job ad clicks.

That 36 applications per hire looks low next to the ATS data below, and that is expected. Appcast only follows candidates who arrived from paid job ads, and it notes its screen to hire numbers cover January to October 2025 because later hires were still in progress.

Appcast’s own costs point to a slightly bigger funnel. Its median cost per hire of $1,053 divided by its median cost per application of $19.20 equals about 55 applications per hire. Medians at each stage do not chain perfectly, so treat 36 to 55 as the paid traffic range.

How many applicants per hire is normal now?

Across all sources of applicants, the number is far higher. Ashby’s productivity report, built on 109 million applications and 247,000 jobs, shows 291 applications per hire in Q1 2026.

The Gem 2026 benchmarks find 0.5% of applicants receive offers, roughly one hire per 200 applications. Smaller companies hire 0.3% of applicants and larger organizations 0.7%.

The Employ 2025 report counts applications per job, not per hire, so it is a near cousin rather than a direct match. It still shows a 24% jump in one year, from 207.2 to 257.5.

Has the number of applicants per hire gone up?

Yes, sharply. Ashby says applications per hire roughly tripled from 2021 to 2024, then eased slightly but stayed near 300 through 2025.

Ashby’s text describes the 2021 starting point as roughly 100 applications per hire, while its chart label shows 121. We use the chart label above.

Are AI generated applications inflating the top of the funnel?

The volume is real and the sources tie it to AI tools. Semafor reported in June 2025 that applications on LinkedIn were up 45% that year, with about 11,000 arriving every minute.

The Greenhouse AI survey from November 2025 found 74% of US job seekers use AI, 49% sent more applications than a year earlier, and 41% admitted trying prompt injections to slip past AI filters.

The same survey found 34% of US recruiters spend up to half their week filtering spam and junk applications. Gem reports applications per recruiter are up 93% compared with 2021.

What the sources do not give is a clean share of applications written by AI. Nobody we checked publishes that number from ATS data, so treat any single percentage you see as a survey estimate at best.

What does more volume do to conversion rates?

It pushes the top stage rate down. Ashby says about 7% to 8% of applications led to an interview in 2021, compared with 4.7% for business roles and 3.6% for technical roles in Q1 2026.

Later stages actually improved. Ashby’s interview to offer rate fell to 5.0% business and 3.9% technical in early 2023, then rose to 10.4% and 7.3% by Q1 2026, above 2021 levels. Better screening at the top means stronger candidates reach interviews.

What is a good offer acceptance rate?

In ATS data, 82% to 84% is normal. Gem’s 2026 report puts it at 82%. The Gem 2025 takeaways gave 84%, against 81% in 2021, so treat Gem’s figure as a low 80s range rather than a precise trend. Employ reports 83.9% in 2025, up from 82.5% in 2024.

One outlier deserves a flag. A Gartner candidate survey found only 48% of candidates accepted their most recent offer in 4Q25, down from 54% in 4Q24 and 85% in 4Q23.

Gartner asks candidates, who may hold several offers at once, while ATS vendors count offers that one employer sent. Both can be true. Gartner did not publish its sample or method in the release.

Acceptance also varies by source. Ashby’s operations benchmarks put median acceptance at 89% for business referrals, 84% for technical referrals and 76% for sourced technical candidates.

Which candidate sources convert best?

Inbound applicants convert worst by far. The Ashby referrals report (January 2021 to December 2024) shows how wide the gap is.

Gem tells the same story. Job boards and company marketing bring about 90% of applications but only about half of hires. Direct sourcing produces 11% of hires from 2.6% of applications, and sourced candidates are nearly 8 times more likely to be hired than inbound applicants.

What reply rate should recruiter outreach get?

About one in six sequences gets a reply. The Gem outreach report, covering 6.2 million sequences sent in 2025, shows reply rates falling year over year.

Follow ups nearly double replies: 8.8% after one message versus 16.3% after three. Gem defines interested rate as sequences that got a reply marked interested.

Does LinkedIn publish an InMail response rate?

Not as a current headline figure. The latest LinkedIn data we could verify, reported in 2022, gives only relative lifts: InMails under 400 characters got 22% more responses than average, and Open to Work candidates responded 37% more.

How do staffing agency funnels differ?

Agencies track submittals, client interviews and placements instead of applications and offers. There is no industry standard benchmark for these ratios. We found no published submittal to interview figure from Bullhorn GRID or Crelate.

The best public data comes from Recruiterflow’s benchmarks, published in July 2026 from a year of data across more than 2,100 recruiting and search firms using its software.

The pieces line up. Multiplying 46% by 33% gives about 15% of submittals placed, or 1 ÷ (0.46 × 0.33) = 6.6 submittals per placement, close to Recruiterflow’s reported 7.

Compare that with Ashby’s agency row above: agency candidates reach interviews 42% of the time but convert to offers only 8%. Clients use agency submittals as a shortlist, then interview widely.

How do I calculate my own recruiting funnel?

Divide the number of candidates who reached a stage by the number who entered the stage before it, for the same set of requisitions and the same time window. Then multiply the stage rates together to get your overall yield.

Split the math by source and by role family. A blended funnel hides the fact that referrals and inbound applicants behave like two different businesses.

Use cohorts, not calendar snapshots. Count what happened to the people who applied in a given month, or your rates will mix candidates from different openings. Our recruiting glossary defines each stage term.

What this means for recruiters

Your biggest leak is at the top, so that is where tooling and process pay off. Faster, fairer screening of inbound volume matters more than squeezing another point out of offer acceptance.

Invest in the channels that convert. Referrals, internal moves and rediscovered past candidates beat job board volume at every stage in both the Ashby and Gem data.

Tie your funnel to money and speed. Plug your stage numbers into the cost per hire calculator, and read our cost per hire benchmarks and time to hire data alongside this page. Screening tools are in our software directory.

For monthly hiring conditions that move these rates, see the Staffing Market Brief.

How we built this page

We pulled every figure from the publisher’s own report, press release or blog post in October 2026 and noted the period each one covers. Ashby data runs January 2021 to March 2026. Gem’s benchmark data runs June 2021 to May 2025. Appcast, Employ and Gem outreach figures cover 2025. CareerPlug covers 2024.

The one hire funnel chart is a Work Insiders model that divides backward through Appcast’s published 2025 median stage rates; the arithmetic is shown in the text. Appcast’s stage rates are listed in stage order in its report, and we mapped them in that order. Where sources disagree, we show each one with its method. We did not find a public Workable, SHRM or Carta funnel benchmark we could verify, so none is used.

Quick answers