1. The ATS is effectively universal — and has been for years
In 2025, Jobscan's annual audit of all 500 Fortune 500 career pages detected an applicant tracking system at 489 of them — 97.8%. The 11 without a detectable ATS likely run proprietary in-house systems rather than none at all. This is not a new development: the figure has hovered between 97.4% and 98.8% every year since 2018.
The practical takeaway is that "should I optimize for ATS?" stopped being a real question years ago. If you apply through an online form at a large employer — and increasingly at mid-size ones — software parses, stores, and indexes your resume before any human sees it.
Fortune 500 companies with a detectable ATS, by year
Jobscan did not publish comparable audits for 2020–2022; the line connects reported years only.
Source: Jobscan 2025 ↗2. Competition per opening has roughly tripled since 2021
Ashby's 2026 Talent Trends Report — built on 109 million applications across 247,000 jobs from January 2021 through March 2026 — found applications per hire tripled from around 100 in early 2021 to more than 300 on average throughout 2025.
The same dataset shows the downstream effect on candidates: a candidate today is roughly 50% less likely to receive an interview than five years ago. Recruiting teams adapted by getting more selective, not slower — hires per recruiter recovered to about seven per quarter, and candidates who do reach interviews convert to offers at rates above 2021 levels.
This volume shift, not a resume-reading robot, is the honest explanation for most of the silence job seekers experience. Being one of 300+ applicants is the default condition of a 2026 job search.
Honest caveat: Ashby's data comes from its own platform, which skews toward technology and startup employers. Volumes at other employer types differ, but the direction of the trend is corroborated across the industry.
Applications per hire, 2021 vs 2025
Reported endpoints from Ashby's platform data (109M+ applications). Intermediate years rose steadily; exact yearly values aren't published, so we chart only the reported endpoints.
Source: Ashby 2026 ↗3. Auto-rejection is mostly myth. Being unfindable is the real risk
In Enhancv's late-2025 study of 25 recruiters across 10+ ATS platforms, 23 of 25 (92%) said their system does not automatically reject resumes based on formatting, content, or design. Only two — using Bullhorn and BambooHR — had configured threshold-based auto-rejection. What 84% do use is knockout questions: binary eligibility filters (work authorization, required licenses, location) set by humans.
That matches what larger research has said for years. Harvard Business School and Accenture's "Hidden Workers" study found more than 90% of employers filter or rank candidates — 94% for middle-skills roles, 92% for high-skills roles — by recruiter-set criteria like credentials and years of experience. The machine executes rules a human wrote.
Meanwhile, Jobscan's recruiter data shows what actually happens instead of auto-rejection: 76.4% of surveyed recruiters search and rank candidates by skills from the job description. A resume that parses badly or misses the searched-for terms isn't rejected — it's simply never found, which feels identical from the outside.
Enhancv recruiter study ↗ · Harvard Business School ↗ · Jobscan 2025 ↗
Honest caveat: The Enhancv study is a small qualitative sample (n=25) — treat its percentages as directional. We include it because it names the systems and settings involved, which larger surveys rarely do.
What recruiters say their systems actually do
Bars come from different surveys — hover or focus each bar for its source and sample caveat.
4. AI screening is real but far from universal — and it makes mistakes
SHRM's State of AI in HR 2026 found 27% of organizations use AI in recruiting — more than any other HR function, but still a minority. Among those organizations, 44% use AI for resume screening, 32% for automating candidate searches, and 29% for applicant communication.
The same research contains the number most coverage skips: 19% of organizations using hiring automation report their tools have screened out qualified candidates. That is a self-reported false-negative rate — the honest version of the "robot rejects good people" story. It happens, it's real, and it's a minority experience rather than the default.
AI on both sides of the hiring table
Hover or focus each bar for its source. The 44% figure is a share of the 27% of organizations already using AI in recruiting, not of all organizations.
5. Candidates are using AI too — the screening arms race is symmetrical
Resume Genius's 2026 survey of 1,000 U.S. job seekers found 38% have used AI tools to help with job applications (25% occasionally, 13% frequently). Among those AI users, 59% use it to write resumes and 48% for cover letters.
Both sides of the hiring table are now algorithmically assisted: a meaningful minority of employers screen with AI, and a meaningful minority of candidates apply with it. The result is more applications per opening (see finding 2), which drives employers toward more filtering — a loop that rewards precise, honest keyword targeting over volume.
6. AI interviews went mainstream in 2026 — and broke candidate trust
Greenhouse's 2026 Candidate AI Interview Report (2,950 job seekers surveyed; US figures n=1,200) found 63% of job seekers have been interviewed by an AI — up 13 percentage points in six months. But 70% were never clearly told upfront that AI would evaluate them, and of candidates who completed an AI interview, 51% never received an outcome at all; only 28% advanced and 13% were formally rejected.
The trust cost is measurable: 38% of candidates have walked away from a hiring process because it included an AI interview. Notably, candidates aren't anti-AI — only 19% want less AI in hiring. What they want is guardrails: upfront disclosure (44%), an explanation of what the AI measures (39%), and the option of a human interview instead (46%).
What happened after candidates completed an AI interview
US respondents (n=1,200) who completed an AI interview. Remainder pending/other.
Source: Greenhouse 2026 ↗7. Which ATS you face depends on who you apply to — and they behave differently
Among Fortune 500 companies, Workday dominates: over 39% use it, with SAP SuccessFactors second at 13.2% — together over half the list. But across the broader market of 12,820 companies in Jobscan's scan data, usage splits: Greenhouse 19.3%, Lever 16.6%, Workday 15.9%, iCIMS 15.3%.
The platforms genuinely differ in what happens after parsing. Greenhouse and Lever don't score or rank resumes — recruiters work from structured scorecards and searches. iCIMS offers an AI "Role Fit" score. Taleo can be configured to disqualify below a threshold. A resume strategy built on one imagined universal "ATS score" mismatches most of the systems actually in use.
ATS usage: Fortune 500 vs the broader market
Fortune 500
Broader market (12,820 companies)
Fortune 500: audit of all 500 career pages (2025). Broader market: 12,820 companies detected in Jobscan scan data.
Source: Jobscan 2025 ↗8. Ghost jobs are a real tax on job seekers
In a Clarify Capital survey of 1,000 hiring managers, roughly one in three admitted their company posts job listings it has no current intention of filling — to build candidate pipelines, signal growth, or keep options open. From the other side, Enhancv's March 2026 survey of 1,000 U.S. job seekers found 47% have applied to a role they later discovered never truly existed, and 25% made it into interview rounds before realizing it.
For candidates this argues for concentrating effort: fewer, better-targeted applications with resumes tuned to the specific posting beat mass-applying into a pool where a meaningful share of listings will never hire anyone.
Clarify Capital ↗ · Enhancv ghost-jobs survey ↗
Honest caveat: Both figures are self-reported surveys, and "ghost job" definitions vary between studies. Treat the exact percentages as soft; the phenomenon itself is corroborated by multiple independent surveys.
Myth check
The number you can safely stop repeating
“75% of resumes are auto-rejected by ATS before a human sees them.”
This statistic traces to a 2012 marketing claim by Preptel, a resume-software vendor that shut down in 2013 without ever publishing a study, sample, or methodology. The figure drifts between 70%, 75%, and 88% depending on who repeats it — a tell that it was never anchored to data. The sourced evidence in findings 3 and 4 describes what actually happens: human-set filters, keyword search, and (in a minority of systems) AI screening.
Full debunk: 10 ATS Myths, each checked against a named source →What this means if you're applying in 2026
Three practical conclusions follow from the sourced data above. First, the battle is findability, not rejection: recruiters overwhelmingly search and rank by skills from the posting, so a resume that parses cleanly and covers the right terms in real context beats any formatting trick. Second, volume rewards targeting: at 300+ applications per hire and with a share of postings never intended to be filled, fewer, sharper, posting-specific applications outperform mass-applying. Third, AI is now on both sides of the table — which raises, not lowers, the value of a resume whose claims survive both machine parsing and the human read that follows.
You can act on all three for free: check how your resume parses with the Resume Health Score, measure it against a specific posting with the Job Match Score, see screening from the recruiter's side in the ATS Simulator, or start from a verified ATS-safe resume template. For the full picture of how these systems work, read How ATS Works.
Beyond the top-line numbers
Two more cuts of the data: how the UK and EU compare to the US picture above, and how the applications-per-hire story looks outside the tech/startup employers most platform data comes from.
9. Outside the US, the adoption data is thinner — but it points the same direction
Nobody runs a Fortune-500-style ATS audit for the UK or EU, so this section is necessarily smaller than the findings above — a fact we'd rather state than paper over. What does exist points the same way as the US data. CIPD's 2024 Resourcing and Talent Planning Report, the UK's largest annual employer survey on hiring practice, found 78% of organisations increased their use of technology in recruitment and onboarding over the prior year, and 31% now use some form of AI or machine learning in recruitment — nearly double the 16% recorded in 2022.
For the wider EU, Eurostat's official enterprise survey doesn't break AI use out by business function, so the closest verifiable figure is broader: 20.0% of EU enterprises with 10 or more employees used AI technologies of any kind in 2025, up from 13.5% in 2024 and 7.7% in 2021 — a near-tripling in four years. That's total business AI use, not recruiting specifically, and we're labelling it that way rather than implying a precision the data doesn't have.
The honest summary: UK and EU employers are adopting hiring technology and AI on a similar trajectory to the US, from a lower current base, without a national- or EU-level equivalent of Jobscan's ATS-vendor audit to say exactly which systems or how many.
Honest caveat: The EU figure measures ALL business AI use (text analysis, image generation, process automation, etc.), not recruiting-specific adoption — no EU-wide recruiting-AI survey comparable to CIPD's or SHRM's exists yet, so we're citing the nearest verifiable official number rather than an unsourced estimate.
AI adoption in hiring/business technology: UK vs EU, by year
Bars come from two different surveys with two different scopes — hover or focus each bar for its source.
UK figures are recruiting-specific (CIPD). EU figures are general business AI use (Eurostat) — no EU-wide recruiting-specific survey exists yet, so the two aren't directly comparable to each other, only to their own prior year.
10. The applications-per-hire gap isn't just tech vs. everyone else
Finding 2 showed applications per hire tripling since 2021 on Ashby's platform, which skews toward tech and startup employers. CareerPlug's 2025 Recruiting Metrics Report — a different population entirely, built from 10M+ applications across 60,000+ small businesses in industries like automotive, retail, healthcare, and restaurants — shows the same volume pressure exists well outside tech, and that it varies enormously by industry even within "small business."
Across all industries in CareerPlug's 2024 data, employers averaged 180 applicants per hire. Automotive employers saw the most at around 234; education and childcare employers saw the fewest at 57. Only about 3% of applicants were invited to interview on average — in line with the same findability problem the core findings above describe.
The takeaway for job seekers: "how many other applicants am I up against" has no single answer. It depends heavily on industry, and matters more than most advice acknowledges when deciding where to invest tailoring effort.
Honest caveat: CareerPlug's customer base is small and local businesses, not large enterprises — a different population from the Ashby (tech/startup) and Jobscan (Fortune 500) data elsewhere on this page. We show it precisely because it's a different population, not to blend the numbers together.
Applications per hire by industry, small businesses (2024)
CareerPlug 2025 Recruiting Metrics Report — 10M+ applications, 60,000+ small businesses, 2024 data. A different, smaller-employer population from the Ashby figures in finding 2.
Source: CareerPlug 2025 ↗Every key statistic, in one citable sentence
Each sentence below is self-contained — the claim and its original source in one quotable line. Reuse any of them freely with a link back to this page; the full context, charts, and honest caveats are in the numbered findings above.
“97.8% of Fortune 500 companies used a detectable applicant tracking system in 2025, according to Jobscan's annual audit of all 500 career pages — a figure that has stayed between 97.4% and 98.8% every year since 2018.”
“Applications per hire roughly tripled between 2021 and 2025 — from about 100 to more than 300 on average — according to Ashby's 2026 Talent Trends Report, built on 109 million applications across 247,000 jobs.”
Full finding, chart, and caveats ↑ · Source: Ashby 2026
“92% of recruiters surveyed by Enhancv in late 2025 (23 of 25, across 10+ ATS platforms) said their applicant tracking system does not automatically reject resumes based on formatting, content, or design — the real risk is being unfindable in recruiter keyword searches, not auto-rejection.”
Full finding, chart, and caveats ↑ · Source: Enhancv recruiter study · Harvard Business School · Jobscan 2025
“27% of organizations use AI in recruiting — more than any other HR function — and 19% of organizations using hiring automation report their tools have screened out qualified candidates, according to SHRM's State of AI in HR 2026.”
Full finding, chart, and caveats ↑ · Source: SHRM 2026
“38% of U.S. job seekers have used AI tools to help with job applications, and among those users 59% use AI to write resumes, according to Resume Genius's 2026 survey of 1,000 U.S. job seekers.”
Full finding, chart, and caveats ↑ · Source: Resume Genius 2026
“63% of job seekers have now been interviewed by an AI — up 13 percentage points in six months — yet 70% were never clearly told upfront that AI would evaluate them, according to Greenhouse's 2026 Candidate AI Interview Report (2,950 job seekers surveyed).”
Full finding, chart, and caveats ↑ · Source: Greenhouse 2026
“Over 39% of Fortune 500 companies use Workday as their ATS, but across the broader market of 12,820 companies in Jobscan's 2025 scan data, usage splits between Greenhouse (19.3%), Lever (16.6%), Workday (15.9%), and iCIMS (15.3%).”
“Roughly one in three hiring managers admit their company posts job listings it has no current intention of filling, according to a Clarify Capital survey of 1,000 hiring managers — and 47% of U.S. job seekers say they have applied to a role they later discovered never truly existed (Enhancv, March 2026, n=1,000).”
Full finding, chart, and caveats ↑ · Source: Clarify Capital · Enhancv ghost-jobs survey
“31% of UK employers use AI or machine learning in recruitment as of 2024 — nearly double the 16% recorded in 2022 — according to CIPD's Resourcing and Talent Planning Report, the UK's largest annual employer survey on hiring practice.”
Full finding, chart, and caveats ↑ · Source: CIPD 2024 · Eurostat 2025
“Even among small businesses, employers averaged 180 applicants per hire in 2024 — ranging from 234 in automotive down to 57 in education and childcare — according to CareerPlug's 2025 Recruiting Metrics Report (10M+ applications, 60,000+ small businesses).”
Full finding, chart, and caveats ↑ · Source: CareerPlug 2025
Cite this study
BookMyJobInterview.ai (2026). The State of ATS 2026: What applicant tracking systems actually do to your job search. https://bookmyjobinterview.ai/state-of-ats-2026
Journalists, career coaches, and bloggers are welcome to reuse any finding or chart from this page with a link back to it as the source. Underlying claims should be attributed to their original sources, all of which are named and linked in the sources section below.
Share a stat
Reusable stat cards from this study — each carries its original source on the image.
These cards are free to reuse — each states its original source. They contain study statistics only: no personal data exists in the images or their URLs, by construction.
Frequently asked questions
Do 75% of resumes really get rejected by an ATS before a human sees them?+
No — that figure isn't sourced to any real study (see the myth check below). What the sourced data shows instead: 92% of surveyed recruiters say their ATS doesn't auto-reject resumes on formatting or content, and most systems are used to search and rank candidates by skills from the posting, not to auto-reject them.
Related finding: no-auto-reject
Does every company use an ATS?+
Almost every large one does. 97.8% of Fortune 500 companies had a detectable applicant tracking system in Jobscan's 2025 audit, a figure that's stayed between 97.4% and 98.8% every year since 2018. Smaller and non-US employers use ATS software too, just less universally and with thinner public data on exact adoption rates.
Related findings: adoption · regional-uk-eu
How many people am I really competing against for one job?+
It depends heavily on the employer and industry. Ashby's platform data (tech/startup-skewed) shows applications per hire roughly tripled since 2021, from about 100 to 300+. CareerPlug's small-business data shows a 2024 average of 180, ranging from 234 in automotive down to 57 in education and childcare. There's no single honest number — only that it's higher than it used to be almost everywhere.
Related findings: volume · employer-size
Is AI actually screening my resume?+
Sometimes, but it's a minority of employers, not the default. SHRM found 27% of organizations use AI anywhere in recruiting, and 44% of those use it specifically for resume screening — meaning roughly 1 in 8 organizations overall. The same research found 19% of organizations using hiring automation have had it screen out qualified candidates, which is real but not the norm.
Related finding: ai-employers
Do UK and EU employers use ATS and AI hiring tools the same way as the US?+
The trend looks similar but the public data is thinner. In the UK, CIPD found 31% of employers use AI/ML in recruitment as of 2024, up from 16% in 2022. Across the EU, Eurostat's broader (non-recruiting-specific) survey found 20.0% of enterprises used AI technologies of any kind in 2025, up from 7.7% in 2021. Both are rising quickly from a lower base than the US Fortune 500 numbers above, which is a fair summary of where the evidence currently stands.
Related finding: regional-uk-eu
Have AI interviews become normal?+
Yes, fast. Greenhouse found 63% of job seekers have now been interviewed by an AI, up 13 points in six months, but 70% weren't clearly told upfront, and 51% of people who completed one never got an outcome at all. Candidates aren't against AI interviews outright — only 19% want less AI in hiring — but they want disclosure and a human-interview option.
Related finding: ai-interviews
Methodology
This study synthesizes publicly available research published between 2021 and mid-2026, selected under three rules: every statistic must trace to a named organization with a stated methodology; sample sizes and data provenance are disclosed inline; and claims that circulate widely but cannot be traced to a methodology (most famously the "75% auto-rejected" figure) are excluded or explicitly debunked rather than repeated. Where sources conflict or cover different populations (e.g., Fortune 500 vs the broader employer market), we present them side by side instead of blending them. Vendor-published research (Jobscan, Ashby, Greenhouse, Enhancv, Resume Genius) is labeled as such; SHRM and Harvard Business School provide the non-vendor anchors. Original sources were accessed and verified on July 18, 2026. This page and PDF will be corrected if any source issues a revision — the version number below changes when the data does.
Version 1.0 · Published 2026-07-18 · Compiled by BookMyJobInterview.ai
Sources
- Jobscan: 2025 Applicant Tracking System (ATS) Usage Report (5th edition) ↗
July 2025
Manual review of all 500 Fortune 500 career pages (list of June 2, 2025) plus ATS-detection data from 1M+ resume scans across 12,820 companies. Vendor-published research.
- Ashby: 2026 Talent Trends Report ↗
May 2026
Analysis of 109M+ applications across 247K+ jobs on the Ashby platform, January 2021 – March 2026. Platform data, skewed toward tech/startup employers.
- Greenhouse: 2026 Candidate AI Interview Report ↗
May 2026
Survey of 2,950 active job seekers (US, UK, Germany, Australia, Ireland). Figures cited here are US respondents (n=1,200) unless noted. Vendor-commissioned survey.
- SHRM: The State of AI in HR 2026 ↗
2026
Survey of HR professionals by the Society for Human Resource Management, the main professional body for HR. Independent of any ATS vendor.
- Enhancv: Does the ATS Reject Your Resume? 25 Recruiters Explain What Really Happens ↗
Sept–Oct 2025
Qualitative study of 25 recruiters across company sizes (100 to 50,000+ employees) using 10+ ATS platforms. SMALL SAMPLE — directional, not statistically representative, and we label it that way.
- Harvard Business School & Accenture: Hidden Workers: Untapped Talent ↗
2021
Joint HBS/Accenture research on hiring filters, surveying 2,250+ executives. Older (2021) but the canonical study on filter-driven screening; nothing comparable has superseded it.
- Resume Genius: 2026 U.S. Job Seekers Survey / Job Search Statistics Report ↗
2026
Survey of 1,000 U.S. job seekers conducted via Pollfish. Vendor-published survey.
- Clarify Capital: Ghost jobs survey ↗
Jan 2025
Survey of 1,000 hiring managers/employers on posting jobs without intent to hire. Self-reported.
- Enhancv: Ghost jobs survey (with BLS data comparison) ↗
March 2026
Survey of 1,000 U.S. job seekers. Self-reported experiences; vendor-published.
- CIPD (Chartered Institute of Personnel and Development): Resourcing and Talent Planning Report 2024 (24th edition) ↗
September 2024
Annual UK employer survey by CIPD, the main professional body for HR, produced with Omni RMS. Independent of any ATS vendor. Figures verified directly on the CIPD report page July 18, 2026.
- Eurostat (European Commission): EU survey on ICT usage and e-commerce in enterprises — use of AI in enterprises ↗
December 2025
Official EU statistical survey of enterprises with 10+ employees across all EU member states. Measures general business AI adoption (text/image/speech tools, machine learning, process automation), NOT recruiting-specific AI or ATS use — the closest thing to an EU-wide, government-run adoption number, cited here with that scope made explicit.
- CareerPlug: 2025 Recruiting Metrics Report: Benchmark Data by Industry ↗
April 2025
CareerPlug's sixth annual report, analyzing 2024 hiring activity from 60,000+ small businesses and 10M+ job applications on its platform. Vendor-published; skews toward small/local employers (retail, automotive, healthcare, restaurants) — a useful counterweight to Ashby's tech/startup-skewed data above.
Where we come in
The data says findability wins: resumes that parse cleanly and surface for the searches recruiters actually run. That is exactly what our service does — we rewrite your resume and LinkedIn profile for your specific target role, verified by an expert ATS reviewer and delivered in 72 hours.