A real number for your role and industry, built from 500,000 sourcing sequences, 54 million applications and US government hiring data. Every rate is cited, and you can open each one to see the source. No sign-up.
Share of sourced candidates who reply to any email in the sequence. Ashby measured a 19.6% average across more than 500,000 sequences. A single email averages 7%; a three-email sequence averages 23%.
Engineering candidates reply at about 21% and are the least likely to open a sourcing email, despite receiving the most outreach.
Metric: Campaign reply rate
Dataset: Over 500,000 sourcing email sequences sent through Ashby, January 2022 to January 2024.
Source: Ashby, Talent Trends Report: Candidate Sourcing (2024)38 of every 182 candidates sourced are still in the process after this step.
Not every reply is a yes. Ashby classified replies and found a 37.7% interest rate, so roughly two in five people who answer are actually open to the role.
Metric: Interest rate
Dataset: Over 500,000 sourcing email sequences sent through Ashby, January 2022 to January 2024.
Source: Ashby, Talent Trends Report: Candidate Sourcing (2024)14 of every 182 candidates sourced are still in the process after this step.
Share of candidates at the recruiter screen who move to the next stage. Referred candidates pass at 52%, so a warm intro roughly halves the work at this step.
Metric: Recruiter screen passthrough
Dataset: 54 million applications across 93,000 jobs, January 2021 to March 2026.
Source: Ashby, Talent Trends Report: Recruiting Operations Benchmarks (2026)5 of every 182 candidates sourced are still in the process after this step.
Share of candidates who get through the main interview loop. This is the tightest step in the funnel and the one that decides how many candidates you need at the top.
Metric: Onsite passthrough
Dataset: 54 million applications across 93,000 jobs, January 2021 to March 2026.
Source: Ashby, Talent Trends Report: Recruiting Operations Benchmarks (2026)1 of every 182 candidates sourced are still in the process after this step.
Share of candidates who clear whatever sits between the loop and an offer, such as a final panel, references or approval. Almost everyone who reaches this point gets through.
Metric: Post-onsite passthrough
Dataset: 54 million applications across 93,000 jobs, January 2021 to March 2026.
Source: Ashby, Talent Trends Report: Recruiting Operations Benchmarks (2026)1 of every 182 candidates sourced are still in the process after this step.
Share of offers that are accepted. Gem reports 82% across 1.2 million hires, so the two large datasets agree closely. Sourced technical candidates accept at 76%, so treat this stage as slightly optimistic for cold outbound.
Metric: Offer acceptance rate
Dataset: 54 million applications across 93,000 jobs, January 2021 to March 2026.
Source: Ashby, Talent Trends Report: Recruiting Operations Benchmarks (2026)1 of every 182 candidates sourced are still in the process after this step.
Openings per hire is how many unfilled roles sit behind each hire an industry actually makes. A high figure means employers are posting faster than they can fill, so each hire takes more sourcing. This is the one number here that is an adjustment rather than a measured sourcing rate, so it is shown separately and you can switch it off.
Without it, the same funnel gives 195 candidates per hire. With it, 195 x 0.93 = 182.
Dataset: Seasonally adjusted hires and job openings by industry, 12 months through July 2026, retrieved from the BLS public API.
Series: JTS510000000000000HIL (hires) and JTS510000000000000JOL (openings), via https://api.bls.gov/publicAPI/v1/timeseries/data/
Source: US Bureau of Labor Statistics, Job Openings and Labor Turnover Survey (JOLTS)Ashby publishes passthrough rates blended across every source, so referrals and internal moves lift them. Gem measures outbound candidates on their own and finds them harder: they clear the first screen easily, because a sourcer already vetted them, then convert less well at the interview and offer stages than a referral would.
Treat 182 as the optimistic end and 471 as the conservative one. If your pipeline is mostly cold outbound, plan closer to the higher number.
Dataset: 165 million applications, 15 million candidates and 1.2 million hires, June 2021 to May 2025.
Source: Gem, 2026 Recruiting Benchmarks Report (PDF)Four first-party datasets, four different answers, and Gem's own two 2026 reports disagree by six points. That is real movement rather than error: reply rates fell through 2025. Read the headline number as the middle of a range, not a precise figure.
23.0% — Gem, June 2024 to May 2025, outbound candidates
19.6% — Ashby, January 2022 to January 2024
16.9% — Gem, calendar 2025
15.2% — hireEZ, January to June 2025
Your own conversion rates always beat a benchmark. Enter any stage as a percentage and the whole calculation follows it. Leave a field empty to keep the published figure.
How many candidates to source: Engineering in Technology, Media & Information
Source 182 candidates per hire.
For 3 hires: 547 candidates.
Pipeline for one hire:
Candidates to source: 182
Replies to your outreach: 38
Reply is positive: 14
Passes the recruiter screen: 5
Passes the interview loop: 1
Clears final checks: 1
Accepts the offer: 1
Rates used:
Replies to your outreach: 21.0% (Campaign reply rate, Ashby, Candidate Sourcing (2024))
Reply is positive: 37.7% (Interest rate, Ashby, Candidate Sourcing (2024))
Passes the recruiter screen: 35.0% (Recruiter screen passthrough, Ashby, Recruiting Operations Benchmarks (2026))
Passes the interview loop: 24.0% (Onsite passthrough, Ashby, Recruiting Operations Benchmarks (2026))
Clears final checks: 95.0% (Post-onsite passthrough, Ashby, Recruiting Operations Benchmarks (2026))
Accepts the offer: 81.0% (Offer acceptance rate, Ashby, Recruiting Operations Benchmarks (2026))
Market adjustment: 0.93x. Technology, Media & Information has 1.27 open roles per hire against a US average of 1.365, from BLS JOLTS data through July 2026.
Cross-check: Gem measures outbound candidates at 2.5% application-to-hire, which puts the same outreach at 471 candidates per hire. Published sequence reply rates span 15.2% to 23.0%, so treat this as a range.
Sources:
Ashby, Talent Trends Report: Candidate Sourcing (2024): https://www.ashbyhq.com/talent-trends-report/reports/candidate-sourcing
Ashby, Talent Trends Report: Recruiting Operations Benchmarks (2026): https://www.ashbyhq.com/talent-trends-report/reports/recruiting-operations-benchmarks-talent-trends
Gem, Key takeaways from the 2026 Recruiting Benchmarks Report (2026): https://www.gem.com/blog/key-takeaways-from-the-2026-recruiting-benchmarks-report
Gem, Email outreach for recruiters: 2026 benchmarks and best practices (2026): https://lp.gem.com/rs/972-IVV-330/images/Email%20Outreach%20Benchmarks%20Report%202026.pdf
hireEZ, 2025 email outreach benchmark report (2025): https://explore.hireez.com/hubfs/eBook-pdf/2025-email-outreach-benchmark-report.pdf
US Bureau of Labor Statistics, Job Openings and Labor Turnover Survey (JOLTS) (2026): https://www.bls.gov/jlt/Over 500,000 sourcing email sequences sent through Ashby, January 2022 to January 2024.
https://www.ashbyhq.com/talent-trends-report/reports/candidate-sourcing54 million applications across 93,000 jobs, January 2021 to March 2026.
https://www.ashbyhq.com/talent-trends-report/reports/recruiting-operations-benchmarks-talent-trends165 million applications, 15 million candidates and 1.2 million hires, June 2021 to May 2025.
https://www.gem.com/blog/key-takeaways-from-the-2026-recruiting-benchmarks-report6.2 million email sequences and 15.5 million messages sent in calendar 2025.
https://lp.gem.com/rs/972-IVV-330/images/Email%20Outreach%20Benchmarks%20Report%202026.pdfOver 2.7 million recruiting emails sent January to June 2025.
https://explore.hireez.com/hubfs/eBook-pdf/2025-email-outreach-benchmark-report.pdfSeasonally adjusted hires and job openings by industry, 12 months through July 2026, retrieved from the BLS public API.
https://www.bls.gov/jlt/“How many candidates do we need?” usually gets answered with a number someone half remembers. The real answer is a chain of conversion rates, and each link has been measured on a large dataset. Source a hundred people and about twenty reply. Fewer than half of those replies are a yes. Of the people who start the process, roughly a third pass the recruiter screen, a quarter get through the interview loop, almost everyone clears final checks, and about four in five accept the offer.
Multiply the chain and you land near 200 sourced candidates per hire for a typical role. That number is not a guess, and it is not ours: it comes from Ashby’s analysis of more than 500,000 sourcing sequences and 54 million applications, cross-checked against Gem’s 1.2 million hires. The calculator shows every link, and you can open each one to see the metric, the dataset and the source.
The stage that varies most is the first one. Ashby measured a 42% reply rate for legal and recruiting candidates, 21% for engineers and 20% for finance, against a 19.6% average. Engineers receive the most sourcing outreach of any group and open the least of it, so an engineering search can need twice the volume of a legal one to produce the same hire. Sequence length matters too: one email averages a 7% reply rate, three emails average 23%, and a fourth adds almost nothing.
Conversion rates describe your process. They say nothing about whether every other employer is chasing the same people. For that the calculator uses the US Bureau of Labor Statistics JOLTS release, which reports actual hires and unfilled openings by industry every month. Dividing openings by hires gives a plain measure of how far demand runs ahead of what the industry can fill.
The tool applies that ratio as a clearly labelled multiplier, shows the raw hires and openings behind it, names the exact BLS series so you can pull the data yourself, and lets you switch it off. It is the one figure that is an adjustment rather than a measured sourcing rate, so it is never folded in silently.
The most useful moment for this number is the kick-off meeting. A hiring manager who wants three engineers this quarter is asking for several hundred sourced candidates, a few hundred outreach replies and a dozen interview loops. Showing that chain, with citations, turns “why is this taking so long?” into a conversation about capacity. Copy the plan out of the tool and paste it into the intake doc.
When you have your own conversion rates, use them. Enter any stage as a percentage and the whole plan follows your data, with the drill-down marking which rates are yours. Then build the search itself with the Boolean Search String Builder, widen it with the Synonym Expander, and point it at public profiles with the X-ray Search Builder.
Knowing you need a few hundred candidates is the easy half. HeroHunt searches 1 billion+ profiles, screens every match against your role and runs the outreach, so the top of the funnel fills itself.
