Statistics · Updated 14 October 2026 · Next review 15 January 2027
ATS Statistics 2026
Applicant tracking systems are nearly universal at large employers, but most of what job seekers hear about them has no source. This page gives each number with who was measured, how and when, separates using an ATS from being rejected by one, and traces the famous "75% auto-rejected" figure back to a 2012 sales pitch.
Key numbers
- 487 of the 2026 Fortune 500 companies (97.4%) had a detectable applicant tracking system behind their career pages. Details
- Workday runs hiring at over 40% of the 2026 Fortune 500, and Workday plus SAP SuccessFactors together cover 52.4%. Details
- 19% of 1,000 US hiring managers said AI screens out some candidates using rules people set, and 6% said AI can advance or reject candidates with little human review. Details
- 99.7% of 384 recruiters surveyed by Jobscan in 2025 said they use filters in their ATS, and 76.4% start by filtering on skills. Details
- In a 2020 survey of 2,275 executives, 88% agreed that qualified high-skills candidates are vetted out because they do not match the exact criteria in the job description, rising to 94% for middle-skills roles. Details
- Between 3.6% and 4.7% of applications in Ashby's ATS led to an interview in early 2026, depending on role type, down from about 7% to 8% in 2021. Details
- About 1% of roughly 200,000 real resumes submitted through the hireEZ hiring platform between 2019 and 2025 contained hidden prompt injections. Details
- 72% of the 257 organisations in Deloitte India's Campus Workforce Trends 2026 study use AI for resume screening, the most common AI use in their campus hiring. Details
- In Mobley v. Workday, Workday told the court that 1.1 billion applications were rejected using its software tools during the period in question. Details
- In a 2024 University of Washington audit, AI text-embedding models favoured resumes with White-associated names in 85.1% of comparisons and female-associated names in only 11.1%. Details
How many employers use an applicant tracking system?
How many Fortune 500 companies use an ATS?
487 of the 2026 Fortune 500 companies (97.4%) had a detectable applicant tracking system behind their career pages.
Jobscan staff checked every company's career pages by hand after the list came out on June 3, 2026; its earlier checks found 97.4% to 98.8% between 2018 and 2025. The other 13 may run in-house systems, and using an ATS says nothing about whether it rejects anyone automatically.
Primary source: Jobscan, 2026 Applicant Tracking System (ATS) Usage Report (August 20, 2026). Jobscan staff manually checked the career pages of all 2026 Fortune 500 companies (list released June 3, 2026) to detect the ATS behind each; companies were not contacted. The wider sample is ATS detections from over 1 million job links pasted into Jobscan's scanner for 13,130 companies since 2016. Jobscan sells resume-optimization software.
Do smaller employers use screening software too?
In a 2020 survey of 2,275 executives in the US, UK and Germany, 63% said their company uses recruiting management software, rising to 75% of US employers.
Use was 69% at firms with more than 1,000 workers, 58% in the UK and 54% in Germany, and the survey included midsize firms of 50 to 999 employees. Harvard Business School and Accenture published it in September 2021.
Primary source: Harvard Business School Project on Managing the Future of Work and Accenture, Hidden Workers: Untapped Talent (September 2021 (corrected October 4, 2021)). Survey of 2,275 executives (at least 750 each in the US, UK and Germany), January to February 2020, plus 8,720 hidden workers in the same countries, May to June 2020.
What share of US hiring managers work with an ATS?
71% of 1,000 US hiring managers in a survey published in January 2026 said their company uses an ATS.
Use was 85% among Gen Z hiring managers and 35% among Baby Boomers. The report's summary says 71% "screen applications" with an ATS while the body says 71% "use" one, so read it as use.
Vendor survey: Resume Genius, 2026 Hiring Insights Report (January 29, 2026 (updated March 24, 2026)). Online survey of 1,000 US hiring managers screened as directly responsible for hiring, via Pollfish Random Device Engagement. Fieldwork dates and margin of error not reported. Resume Genius sells resume tools.
How many organizations use AI in recruiting?
Recruiting was the HR area with the highest AI adoption in SHRM's 2026 survey of 1,908 HR professionals, at 27% of organizations.
Seven of the top 10 AI uses in HR were in recruiting, including resume screening, job posting creation and candidate matching. Larger employers lead: 60% of extra-large organizations had adopted AI in HR, against about 1 in 3 small ones.
Primary source: SHRM, The Real Impact of AI in HR: March 2026 EN:Insights Forum recap (The State of AI in HR 2026) (March 30, 2026 (full report March 31, 2026)). Survey of 1,908 HR professionals for SHRM's State of AI in HR 2026 report, as previewed by SHRM's lead HR technology analyst; fieldwork dates not stated on the page.
Do Indian employers use AI to screen resumes?
72% of the 257 organisations in Deloitte India's Campus Workforce Trends 2026 study use AI for resume screening, the most common AI use in their campus hiring.
AI-based assessments (48%) and candidate engagement (37%) came next. The figures come from press coverage of the report, Deloitte has not published the survey period, and the sample is large campus recruiters, not all Indian employers.
Reported by a secondary source: SightsIn Plus (reporting Deloitte India), Deloitte releases Campus Workforce Trends 2026 report (Report released August 27, 2026; article August 31, 2026). Deloitte India study of 257 organisations and 505 campuses (MBA, engineering, BBA, B.Com, B.Pharma, law and other programmes) across manufacturing, services, pharma, consumer, financial services and technology; survey period not published.
Which systems do employers use?
How do recruiters in India find candidates?
Recruiters ran about 9.97 lakh (997,000) resume searches a day on Naukri in April to June 2026.
These are recruiters searching Naukri's 118 million stored resumes by keyword, a different route from applying through an employer's ATS. Company-reported daily average for Q1 FY27 from Info Edge's investor presentation.
Primary source: Info Edge (India) Ltd, Investor presentation, quarter ended June 30, 2026 (Naukri.com metrics) (August 2026). Company-reported operating metrics for Naukri.com as of June 30, 2026 or averaged over April to June 2026 (Q1 FY27), from the listed parent's quarterly investor presentation.
What does an ATS actually do with your resume?
How do recruiters search inside an ATS?
99.7% of 384 recruiters surveyed by Jobscan in 2025 said they use filters in their ATS, and 76.4% start by filtering on skills.
Other common filters were education (59.7%), job title (55.3%), certifications or licences (50.6%), years of experience (44%) and location (43.4%). A filter narrows what a recruiter sees in a search, which is different from an automatic rejection.
Vendor survey: Jobscan, State of the Job Search 2025 (February 25, 2025). Online survey of 384 randomly sampled HR professionals directly involved in hiring, February to March 2025, via Zoho Surveys. Jobscan sells resume-optimization software.
Do ATS filters screen out employment gaps?
In a 2020 survey of 2,275 executives, 48% said their recruiting software filtered out middle-skills candidates with an employment gap of more than six months.
More than 90% said they used such software to filter or rank candidates at the first screen (94% for middle-skills and 92% for high-skills roles). The report notes that applicants removed by a rule like this are never seen by a recruiter.
Primary source: Harvard Business School Project on Managing the Future of Work and Accenture, Hidden Workers: Untapped Talent (September 2021 (corrected October 4, 2021)). Survey of 2,275 executives (at least 750 each in the US, UK and Germany), January to February 2020, plus 8,720 hidden workers in the same countries, May to June 2020.
What does an AI match score inside an ATS do?
Greenhouse's Talent Matching sorts applicants into 4 match tiers (Strong, Good, Partial, Limited) and, per its documentation, does not auto-reject or auto-advance anyone.
If it cannot read a resume, the applicant is labelled "Needs manual review" for a person to check. This describes one vendor's feature from its own documentation; other ATS work differently.
Primary source: Greenhouse, Talent Matching FAQ (Last updated October 5, 2026). Vendor product documentation describing how Greenhouse's AI matching feature labels applicants and what it does when a resume cannot be read; not an independent test.
Are job applications asking more questions?
The number of questions per job application nearly doubled between 2021 and 2024, across more than 4.8 million applications in Ashby's ATS.
In 2024, 46.8% of application forms included a long-form question. Ashby did not break out knockout questions such as work authorization, and the data covers one vendor's customers.
Primary source: Ashby, Are jobs asking more application questions? (March 18, 2025). Analysis of more than 4.8 million submitted applications in Ashby's ATS, January 2021 to December 2024; abandoned applications are not counted. Data from one vendor's customers.
Can AI resume screeners be biased?
In a 2024 University of Washington audit, AI text-embedding models favoured resumes with White-associated names in 85.1% of comparisons and female-associated names in only 11.1%.
Researchers changed only the name on over 500 resumes matched to 500 job descriptions in nine occupations, and Black male names were disadvantaged in up to 100% of comparisons. The models were open-source and tested in a lab, not a commercial ATS in use.
Primary source: AAAI/ACM Conference on AI, Ethics, and Society (AIES 2024), Gender, Race, and Intersectional Bias in Resume Screening via Language Model Retrieval (October 16, 2024). Audit of Massive Text Embedding models used to retrieve resumes for job descriptions: over 500 public resumes and 500 job descriptions across nine occupations, with only the name changed to signal race and gender. University of Washington; peer-reviewed.
Do employers publish audits of their AI hiring tools?
Of 391 employers checked under New York City's Local Law 144, only 18 had posted a bias audit and 13 a transparency notice.
The law lets employers decide whether their tool is covered, so a missing audit is not proof of a violation; the authors call this "null compliance". Published at ACM FAccT 2024.
Primary source: ACM FAccT 2024 (arXiv preprint), Null Compliance: NYC Local Law 144 and the Challenges of Algorithm Accountability (June 3, 2024). 155 student investigators recorded how 391 employers hiring in New York City complied with Local Law 144, which requires bias audits and notices for automated employment decision tools.
How often does a human see your application?
How many employers let AI reject applicants before a person looks?
19% of 1,000 US hiring managers said AI screens out some candidates using rules people set, and 6% said AI can advance or reject candidates with little human review.
32% said AI ranks candidates but people make every final call, and 22% use AI only for admin. Separately, 37% said they let their ATS screen out applicants on criteria they set, so rule-based filtering is more common than AI making the call.
Vendor survey: Resume Genius, 2026 Hiring Insights Report (January 29, 2026 (updated March 24, 2026)). Online survey of 1,000 US hiring managers screened as directly responsible for hiring, via Pollfish Random Device Engagement. Fieldwork dates and margin of error not reported. Resume Genius sells resume tools.
Do recruiters say their ATS auto-rejects resumes?
23 of 25 US recruiters interviewed by Enhancv in September and October 2025 said their system does not auto-reject resumes for formatting or content.
The other 2 used match thresholds or skill-based auto-rejection, and knockout questions were mostly limited to eligibility such as work authorization or licences. Twenty-five interviews are qualitative and not a representative sample.
Vendor survey: Enhancv, Does the ATS Reject Your Resume? 25 Recruiters Explain What Really Happens (November 3, 2025). Structured interviews with 25 US recruiters and talent acquisition staff, September to October 2025, recorded and coded by theme; recruitment method not described and not representative. Enhancv sells a resume builder.
How sure are recruiters that their systems are not rejecting good candidates?
Only 21% of US recruiters in Greenhouse's 2025 survey said they were very confident their systems are not rejecting qualified candidates.
25% were not confident in their AI systems at all, and 8% said they do not know what their algorithms prioritize. Greenhouse sells an ATS and did not publish its research firm or fieldwork dates.
Vendor survey: Greenhouse, An AI trust crisis: 70% of hiring managers trust AI to make faster and better hiring decisions, only 8% of job seekers call it fair (November 19, 2025). Greenhouse 2025 AI in Hiring Report: survey of 4,136 people in the US, UK, Ireland and Germany (2,900 job seekers and 1,236 recruiters and hiring managers; US: 1,200 job seekers and 665 recruiters and hiring managers). Research firm and fieldwork dates not stated. Greenhouse sells an ATS.
Do employers think screening filters out qualified people?
In a 2020 survey of 2,275 executives, 88% agreed that qualified high-skills candidates are vetted out because they do not match the exact criteria in the job description, rising to 94% for middle-skills roles.
The question was about rigid job criteria across the whole screening process, not about ATS software rejecting resumes on its own. The executive survey ran in January and February 2020, before the pandemic.
Primary source: Harvard Business School Project on Managing the Future of Work and Accenture, Hidden Workers: Untapped Talent (September 2021 (corrected October 4, 2021)). Survey of 2,275 executives (at least 750 each in the US, UK and Germany), January to February 2020, plus 8,720 hidden workers in the same countries, May to June 2020.
How many applications does it take to fill a job?
Roles in Ashby's ATS averaged 291 applications per hire in early 2026, against roughly 100 to 121 in early 2021.
Applications per hire peaked at 319 in Q4 2024; Ashby's text and chart give slightly different 2021 baselines. The data covers more than 109 million applications and 247,000 jobs at one vendor's customers, January 2021 to March 2026.
Primary source: Ashby, Recruiter Productivity: 2026 Talent Trends Report (April 28, 2026). Analysis of more than 109 million applications and 247,000 jobs in Ashby's ATS, January 2021 to March 2026. Data from one vendor's customers.
What are the odds that an application leads to an interview?
Between 3.6% and 4.7% of applications in Ashby's ATS led to an interview in early 2026, depending on role type, down from about 7% to 8% in 2021.
Ashby says candidates are roughly 50% less likely to get an interview than five years ago, as volume per job has risen. Same dataset of more than 109 million applications.
Primary source: Ashby, Recruiter Productivity: 2026 Talent Trends Report (April 28, 2026). Analysis of more than 109 million applications and 247,000 jobs in Ashby's ATS, January 2021 to March 2026. Data from one vendor's customers.
How many applications have been rejected through Workday?
In Mobley v. Workday, Workday told the court that 1.1 billion applications were rejected using its software tools during the period in question.
It is a count of rejections processed through Workday software, not proof that AI made those decisions. On May 16, 2025 the court let an age-discrimination claim over Workday's AI recommendation features proceed as a nationwide collective action, and the claims had not been decided on the merits as of mid-2026.
Reported by a secondary source: Proskauer Rose LLP, Law and the Workplace, AI Bias Lawsuit Against Workday Reaches Next Stage as Court Grants Conditional Certification of ADEA Claim (June 11, 2025). Law-firm summary of the May 16, 2025 order in Mobley v. Workday, Inc., No. 23-cv-00770-RFL (N.D. Cal.), quoting Workday's own court filing. Primary: the court order.
Are candidates trying to trick the ATS, and does it work?
How many job seekers say they have hidden text in their resume?
41% of US job seekers in Greenhouse's 2025 survey admitted using prompt injections, hidden text meant to get past AI filters.
That self-reported share is far above the roughly 1% of real resumes found to contain hidden text. The gap may come from question wording, which Greenhouse did not publish.
Vendor survey: Greenhouse, An AI trust crisis: 70% of hiring managers trust AI to make faster and better hiring decisions, only 8% of job seekers call it fair (November 19, 2025). Greenhouse 2025 AI in Hiring Report: survey of 4,136 people in the US, UK, Ireland and Germany (2,900 job seekers and 1,236 recruiters and hiring managers; US: 1,200 job seekers and 665 recruiters and hiring managers). Research firm and fieldwork dates not stated. Greenhouse sells an ATS.
Which popular numbers have no real source?
"75% of resumes are auto-rejected by an ATS before a person sees them."
No study behind it: it traces to a resume-software vendor's sales pitch around 2012.
Preptel, which sold resume-optimization software and shut down in 2013, never published data or a method, and the number spread through press coverage and resume blogs, sometimes as 70% or 88%. Resume Genius's 2026 survey of 1,000 US hiring managers points the other way: 19% let AI screen out some candidates and 6% let it reject with little human review. (UnchartedCareer)
"Almost every big company uses an ATS, so software decides who gets an interview."
Adoption is real; automatic decisions are not the norm.
An ATS is mostly a database and workflow tool that recruiters search, filter and rank inside. In Enhancv's 2025 interviews, 23 of 25 recruiters said theirs does not auto-reject on content, and the automatic rules they did use were mostly knockout questions about eligibility. (Enhancv)
"An ATS cannot read PDF resumes."
Outdated: a text-based PDF was the format hiring managers most often called ATS-friendly.
53% of US hiring managers named a text-based PDF, ahead of .docx (43%). What breaks parsing is a PDF that is really an image, or a heavily designed layout, which only 13% called ATS-compatible. (Resume Genius)
"Hiding keywords in white text gets you past the ATS."
Never shown to work, and the hidden text shows up in the extracted resume.
About 1% of 200,000 real resumes contained hidden prompts, but the researchers chose not to test whether they changed outcomes. Separately, 22% of US hiring managers in Greenhouse's 2025 survey said they had caught hidden prompt injections. (USENIX Security Symposium 2026 (arXiv preprint))
"88% of employers admit their ATS rejects qualified candidates."
A misquote of a 2020 survey about rigid job criteria, not about software.
The Harvard Business School and Accenture report found 88% of executives agreed that qualified high-skills candidates are vetted out because they do not match the exact criteria in the job description. That describes how employers write requirements and filter on them across the whole process; it is not a measured rejection rate for ATS software. (Harvard Business School Project on Managing the Future of Work and Accenture)
"Workday's AI automatically rejected 1.1 billion job applications."
The 1.1 billion is Workday's count of rejections processed through its software, not of AI decisions.
Workday gave the figure in Mobley v. Workday when the court was weighing how large the collective of applicants could be. Whether its AI features discriminated is what the lawsuit is about, and that had not been decided as of mid-2026. (Proskauer Rose LLP, Law and the Workplace)
"Your ATS match score decides whether you get an interview."
Most recruiters in a 2025 interview study ignored AI fit scores or did not have them.
Of 25 US recruiters, 44% said their ATS shows an AI or fit score; 56% ignore it or do not have one, 36% use it as a rough guide and 8% treat it as decisive. Scores shown by resume-checking websites are those sites' own metrics, not the employer's. (Enhancv)
How do I cite this page?
Use the citation below and link to this page. If you quote a number, please also name its original source, listed with each number.
Resumity, "ATS Statistics 2026", updated 14 October 2026, https://www.resumity.org/ats-statisticsIs your own resume readable by software?
The free check reads your PDF in your browser (the file is not uploaded) and shows whether your name, contact details, headings and dates can be extracted.
Where do these numbers come from?
Every source below was opened and checked on 14 October 2026. Each entry says who was asked or measured, how many, and when.
- Jobscan: 2026 Applicant Tracking System (ATS) Usage Report (August 20, 2026). Jobscan staff manually checked the career pages of all 2026 Fortune 500 companies (list released June 3, 2026) to detect the ATS behind each; companies were not contacted. The wider sample is ATS detections from over 1 million job links pasted into Jobscan's scanner for 13,130 companies since 2016. Jobscan sells resume-optimization software.
- Harvard Business School Project on Managing the Future of Work and Accenture: Hidden Workers: Untapped Talent (September 2021 (corrected October 4, 2021)). Survey of 2,275 executives (at least 750 each in the US, UK and Germany), January to February 2020, plus 8,720 hidden workers in the same countries, May to June 2020.
- Resume Genius: 2026 Hiring Insights Report (January 29, 2026 (updated March 24, 2026)). Online survey of 1,000 US hiring managers screened as directly responsible for hiring, via Pollfish Random Device Engagement. Fieldwork dates and margin of error not reported. Resume Genius sells resume tools.
- SHRM: The Real Impact of AI in HR: March 2026 EN:Insights Forum recap (The State of AI in HR 2026) (March 30, 2026 (full report March 31, 2026)). Survey of 1,908 HR professionals for SHRM's State of AI in HR 2026 report, as previewed by SHRM's lead HR technology analyst; fieldwork dates not stated on the page.
- SightsIn Plus (reporting Deloitte India): Deloitte releases Campus Workforce Trends 2026 report (Report released August 27, 2026; article August 31, 2026). Deloitte India study of 257 organisations and 505 campuses (MBA, engineering, BBA, B.Com, B.Pharma, law and other programmes) across manufacturing, services, pharma, consumer, financial services and technology; survey period not published.
- Info Edge (India) Ltd: Investor presentation, quarter ended June 30, 2026 (Naukri.com metrics) (August 2026). Company-reported operating metrics for Naukri.com as of June 30, 2026 or averaged over April to June 2026 (Q1 FY27), from the listed parent's quarterly investor presentation.
- Jobscan: State of the Job Search 2025 (February 25, 2025). Online survey of 384 randomly sampled HR professionals directly involved in hiring, February to March 2025, via Zoho Surveys. Jobscan sells resume-optimization software.
- Greenhouse: Talent Matching FAQ (Last updated October 5, 2026). Vendor product documentation describing how Greenhouse's AI matching feature labels applicants and what it does when a resume cannot be read; not an independent test.
- Ashby: Are jobs asking more application questions? (March 18, 2025). Analysis of more than 4.8 million submitted applications in Ashby's ATS, January 2021 to December 2024; abandoned applications are not counted. Data from one vendor's customers.
- AAAI/ACM Conference on AI, Ethics, and Society (AIES 2024): Gender, Race, and Intersectional Bias in Resume Screening via Language Model Retrieval (October 16, 2024). Audit of Massive Text Embedding models used to retrieve resumes for job descriptions: over 500 public resumes and 500 job descriptions across nine occupations, with only the name changed to signal race and gender. University of Washington; peer-reviewed.
- ACM FAccT 2024 (arXiv preprint): Null Compliance: NYC Local Law 144 and the Challenges of Algorithm Accountability (June 3, 2024). 155 student investigators recorded how 391 employers hiring in New York City complied with Local Law 144, which requires bias audits and notices for automated employment decision tools.
- Enhancv: Does the ATS Reject Your Resume? 25 Recruiters Explain What Really Happens (November 3, 2025). Structured interviews with 25 US recruiters and talent acquisition staff, September to October 2025, recorded and coded by theme; recruitment method not described and not representative. Enhancv sells a resume builder.
- Greenhouse: An AI trust crisis: 70% of hiring managers trust AI to make faster and better hiring decisions, only 8% of job seekers call it fair (November 19, 2025). Greenhouse 2025 AI in Hiring Report: survey of 4,136 people in the US, UK, Ireland and Germany (2,900 job seekers and 1,236 recruiters and hiring managers; US: 1,200 job seekers and 665 recruiters and hiring managers). Research firm and fieldwork dates not stated. Greenhouse sells an ATS.
- Ashby: Recruiter Productivity: 2026 Talent Trends Report (April 28, 2026). Analysis of more than 109 million applications and 247,000 jobs in Ashby's ATS, January 2021 to March 2026. Data from one vendor's customers.
- Proskauer Rose LLP, Law and the Workplace: AI Bias Lawsuit Against Workday Reaches Next Stage as Court Grants Conditional Certification of ADEA Claim (June 11, 2025). Law-firm summary of the May 16, 2025 order in Mobley v. Workday, Inc., No. 23-cv-00770-RFL (N.D. Cal.), quoting Workday's own court filing. Primary: the court order.
- USENIX Security Symposium 2026 (arXiv preprint): Measuring Real-World Prompt Injection Attacks in LLM-based Resume Screening (May 27, 2026). Analysis of about 200,000 real resumes collected by the hireEZ hiring platform from July 2019 to December 2025, using purpose-built detectors checked against a hand-labelled sample. Researchers from Duke, UNC Chapel Hill, UC Berkeley and others; outcomes deliberately not tested.
- UnchartedCareer: The "75% of resumes are auto-rejected" myth, traced to its source (Updated July 4, 2026). Source trace of the 75% claim through press coverage back to Preptel, a resume-software vendor that pitched it around 2012 and shut down in August 2013, checked against the HBS and Accenture report and Greenhouse documentation.