Time to Hire in 2026: Benchmarks, Definitions and Where the 35 Days Go

The median time to hire in EMEA is 35 days, measured from the day an application is received to the day the hire is made, according to Ashby's EMEA Recruiter Productivity report (data to June 2026, more than 180,000 EMEA jobs); the AMER median is 34 days. Time to fill is longer because it starts when the requisition opens: SHRM's 2025 recruiting benchmark puts the median at roughly a month and a half for both executive and nonexecutive roles. If you want to reduce either number, the days to recover sit between the stages, in scheduling, feedback and offer turnaround, not inside the interviews.

What is the average time to hire in 2026?

It depends on which definition and which dataset you read, which is why published figures range so widely. The sources I trust most, because they publish their method and their sample:

• Ashby, EMEA Recruiter Productivity (September 2026). Median time to hire over the past year: 35 days in EMEA, 34 in AMER. About 59% of EMEA hires happen within 40 days, and the two regions converge at the 75th percentile by day 50. Since 2021 the EMEA figure has moved within a 37-43 day band. Ashby measures "the average number of days between when an application is received and when a hire is made" and excludes time to start, which it notes runs longer in Europe. Source.
• Ashby, Recruiter Productivity (April 2026, 109M+ applications, 247K+ jobs). The same metric split by role type for Q2-Q3 2025: business roles a median of 30 days (75th percentile 46), technical roles a median of 40 days (75th percentile 57). The report's most recent averages from application to offer are about 38 days for business roles and about 48 for technical (its chart runs to Q1 2026). Source.
• SHRM, State of Recruiting 2025 (more than 2,300 SHRM members). Median time to fill, from job requisition to offer acceptance, "roughly a month and a half" for both executive and nonexecutive positions. In 2017 and 2022 executive roles took 16-17 days longer than nonexecutive ones; that gap has closed. Source.

So a reasonable planning number for 2026: 30-40 days time to hire depending on role type, around a month and a half time to fill. Figures far outside that are a different metric, a different role mix, or one vendor's customer base.

One more number from the Ashby EMEA data before you benchmark yourself: applications per hire have roughly doubled since 2021 (business roles 170 in EMEA vs 229 in AMER, technical roles 254 vs 610), and only 6.7% of applications reached an interview in Q1 2026. More applications per hire means more screening before the first interview, which is one reason time to hire has not fallen even as scheduling tools improved.

Time to fill vs time to hire vs time to submit: which number are you looking at?

Most arguments about "our time to hire is 60 days" are arguments between people using different clocks. Three definitions, with the sources that set them:

Metric	Clock starts	Clock stops	Who uses it	Source
Time to fill	Job requisition opened	Offer accepted, counted in calendar days including weekends and holidays	In-house talent teams, finance	SHRM, Staffing Metrics; SHRM, State of Recruiting 2025
Time to hire	Candidate identified (SHRM) or application received (Ashby)	Offer accepted (SHRM; Ashby Recruiter Productivity) or hire made (Ashby EMEA)	Recruiting operations, ATS reports	SHRM, Data-Driven Recruiting; Ashby
Time to submit	Mandate confirmed	First shortlist sent to the client	Recruitment agencies	No public benchmark found; a working definition from our agency practice

Time to fill is always the larger number, because it includes the sourcing period before anyone is in the pipeline, so SHRM's month and a half and Ashby's 35 days do not contradict each other. And even time to hire has two end points in circulation: SHRM stops at offer acceptance, Ashby's EMEA report at "hire made". If your ATS stops the clock at offer sent, your number will look better than either benchmark. Pick one definition, write it down, and never compare across definitions.

For agencies, time to submit is the number the client actually experiences: how long until the first profiles arrive, then how long each interview takes to schedule. I could not find a public benchmark for it, so we set our own, and I would treat any vendor's "industry average" for it with suspicion.

Where do the 35 days actually go?

Ashby's Recruiting Operations Benchmarks (May 2026) puts it plainly: "Time to Hire is not driven by a single bottleneck, but by a combination of smaller delays across each stage." Nobody has a 35-day process because one thing is broken. They have it because five things each lose a few days.

Here is the 35-day EMEA median reconstructed from the stage-level figures the public reports do give. The "days" column is arithmetic on published medians, not a measurement of any single company. The targets are ours. Figures as published on the linked pages, accessed 2026-10-14.

Stage	What the public data shows	Days (derived)	Our working target	Source
Application to first interview	Only 9% of candidates get a first interview within a day of applying; the largest group (31%) waits 2-3 weeks	7-14	First human response within 2 business days; first interview within 7 calendar days	Cronofy 2024 (12,000 candidates, 7 countries)
Scheduling each interview round	Median about 6 days from availability request to interview; 65% within a week; 8-10% take more than two weeks	6 per round	Booking link sent the moment a candidate advances; slot confirmed within 48 hours	Ashby, Recruiting Coordination (Oct 2024)
Interview loop, first to last interview	Median about 14 days for business roles, 17.9 for technical; 4-5 interview events per hire	14-18	At most 4 events inside 10 calendar days; panel interviews on one day	Ashby, Recruiting Operations Benchmarks; Ashby, Recruiter Productivity
Interviewer feedback and decision	No public dataset isolates it; it hides inside the loop number above	inside the loop	Scorecard submitted within 24 hours of the interview, debrief within 48	Not stated in any report opened
Offer turnaround	Slowest-quartile companies spend 3.26 (business) to 4.71 (technical) more days here than the fastest quartile; 60% of candidates get an offer within a week of the final interview	3-7	Verbal offer within 3 business days of the final interview, written within 5	Ashby, Recruiting Operations Benchmarks; Survale 2025 CandE (66,000+ surveys)

Add the first-interview wait, one interview loop and a normal offer turnaround and you are at 24-39 days before any feedback lag, counter-offer or extra round. The 35-day median is not a mystery. It is the arithmetic of ordinary waiting.

Three things stand out. The interview loop is the largest block, and it is mostly empty: Ashby's Recruiter Productivity report puts total interview time at about 12.2 hours per business hire and 23.3 per technical hire, and twelve hours spread over 14 days is a loop that is idle most of the time. Scheduling appears twice, before the first interview and again inside every round; a 6-day median multiplied by 4-5 events is where most of the loop goes. Offer turnaround is the cheapest fix: the slowest quartile loses 3-5 days here against the fastest, which is approval latency and nothing else, and Survale's data has 40% of candidates not receiving an offer within a week of the final interview.

Why does scheduling eat the most days?

Because it is a coordination problem across people who are not paid to solve it. Yello's recruiter survey (200 recruiting professionals, September 2019) found 35% naming interview scheduling the most time-consuming part of recruiting, 67% spending between 30 minutes and two hours to schedule a single interview, and 60% sometimes or often losing candidates before an interview is scheduled at all.

The Ashby coordination data shows the mechanics. Half of candidates return availability within 24 hours, so the candidate is rarely the slow side. Panel interviews are scheduled within one day in only 34% of cases, against 54% for single interviews. Self-scheduling moves the number: with a direct booking link, 72% of single-event interviews are booked within a day of the availability request, against 54% without, and interviews get scheduled 33% sooner. Automated scheduling confirms interviews in a median 3.7 hours versus 5 hours manually, per the Operations Benchmarks report. It requires the link to go out the moment the candidate advances, not when the recruiter next clears their inbox.

How long do candidates wait before they leave?

Long enough that a 35-day process already loses people. Cronofy's 2024 survey of 12,000 candidates across the UK, France, Germany, the US, Spain, Canada and the Netherlands found that 42% had left a process because scheduling an interview took too long (down from 49% the previous year), 12% would disengage after one week, and 36% at the one-month mark, up on the previous year. The expectations are not unreasonable: 34% of candidates expect a first interview within 2-3 weeks of applying and 29% expect it within about a week.

Survale's 2025 CandE research adds the back end: 31% of North American candidates reported hearing nothing 1-2 months or more after applying, and 56% of rejected candidates got no feedback. EMEA now has the highest "resentment rate" of any region at 17%, meaning candidates who say they will not engage with the employer again after a poor experience, and 2025 is the first year EMEA exceeded North America on it.

The practical reading: the silence count matters more than the day count. A 35-day process with a message every three days loses fewer people than a 25-day process with two weeks of nothing in the middle. What those lost candidates cost is in the real cost of a slow hiring process.

How do you reduce time to hire without cutting evaluation?

Take the days out of the gaps, not out of the interviews. In order of return for effort, based on the table above:

1. Set a scheduling SLA and automate the first step. Booking link at stage advance, 48-hour slot window, one panel day for multi-interviewer rounds. The Ashby data gives you scheduling a third sooner from the link alone.
2. Cap the loop. Four interview events, ten calendar days, written into the brief before sourcing starts. Every extra round is another 6-day scheduling median.
3. Put a clock on feedback. Scorecards within 24 hours, debrief within 48. This is the stage no public dataset measures, so nobody manages it.
4. Pre-approve the offer. Compensation band and approvers agreed when the role opens, so the offer is a form to fill rather than a chain to start. That is the 3-5 day quartile gap.
5. Measure time in stage, not time to hire. One total hides the broken stage. This is the measurement problem in the recruiting analytics nobody measures.

What not to do: drop the structured interview to save a round. In Survale's data 65% of North American employers use structured interviews, and 71% of CandE award winners do; the winners are not faster because they evaluate less.

From our work at ScalingPPL, the stage that surprises clients most in their own data is feedback, not scheduling. The interview happened; the decision did not. Nobody experiences an unmade decision as a delay, because nobody is doing anything during it.

What should an agency measure instead?

An agency cannot control the client's loop, but it owns three clocks and should report all three:

• Time to submit: mandate confirmed to first shortlist sent. Your sourcing speed.
• Time to client feedback: shortlist sent to client decision, per candidate. The client's speed, and the number that justifies a frank conversation when it slips.
• Time to interview: client says yes to interview held. Shared, and where the booking link does most of the work.

Reporting the second one separately matters because, in the Ashby EMEA data, agencies account for about 5% of hires and inbound for 54.3%. An agency candidate waiting two weeks for client feedback is usually also in someone else's inbound pipeline. The client's slowness is your loss, and you can only argue it with a number. If you do not yet know which of your stages is slow, run a recruiting process audit first; most teams guess wrong.

How do you track it in an ATS?

Any ATS reports a total time to hire. The useful report is time in stage per job and per client, with stage-by-stage conversion next to it, so the 35 days decompose into the table above for your own pipeline. That is what Pickr's recruiting analytics do, with a weekly email summary so the number reaches the team without a dashboard. On the process side, Pickr runs automations when a candidate changes stage (confirmation email, calendar invite, a Slack message to the hiring manager), flags candidates who have sat in a stage too long in a daily brief, drafts scorecards from the interview transcript and nudges interviewers while a scorecard is outstanding. Those map onto the gaps in the table.

Pickr is one option. Ashby's own reporting, which produced most of the data above, is extensive, and an in-house team already running it should start there. Whatever the tool, the discipline is the same: one definition, one clock per stage, a target per stage, and a weekly look at which stage missed it.

Sources

• Ashby, EMEA Recruiter Productivity, Talent Trends Report (September 2026), accessed 2026-10-14
• Ashby, Recruiter Productivity, 2026 Talent Trends Report (April 2026), accessed 2026-10-14
• Ashby, Recruiting Operations Benchmarks, 2026 Talent Trends Report (May 2026), accessed 2026-10-14
• Ashby, Recruiting Coordination, Talent Trends Report (October 2024), accessed 2026-10-14
• SHRM, The State of Recruiting 2025: Insights to Maximize Recruitment (October 2025), accessed 2026-10-14
• SHRM, Staffing Metrics: "Time to Fill" Can Kill Prospects of Landing Top Talent (2012), accessed 2026-10-14
• SHRM, Data-Driven Recruiting Proves Business Impact (October 2025), accessed 2026-10-14
• Survale, 2025 Candidate Experience (CandE) Benchmark Research: Fact Sheet and Key Statistics, accessed 2026-10-14
• Cronofy, Candidate Expectations and Preferences 2024, accessed 2026-10-14
• Yello, 2020 Interview Scheduling Statistics and Trends: A Nationwide Recruiter Survey, accessed 2026-10-14