Does ATS reject my resume? Almost never. In interviews with 25 US recruiters, 23 said their systems do not auto-reject on formatting, design, or content. The software ranks and sorts. People decide.
Key Takeaways
- Enhancv’s interviews with 25 US recruiters across 10+ platforms found 92% said their applicant tracking system does not auto-reject resumes on formatting, design, or content.
- The famous claim that 75% of resumes are rejected by software traces to a 2012 Preptel sales pitch; the company closed in 2013 and never published a methodology.
- 44% of those recruiters had an AI fit score available, but only 8% used it as a hard filter.
- The one genuinely automatic gate is the knockout question, covering things like work authorisation or a required licence.
- Parsing failures are real even though rejection is not: Workday strips headers and footers during extraction, so contact details placed there vanish from the parsed profile.
- ResumeAdapter’s Q1 2026 pipeline data puts the median unoptimised fit score at 48 out of 100, with 51% of resumes scoring below 50. That is a queue position, not a verdict.
- 52% of recruiters told Enhancv that applying within the first 48 to 72 hours meaningfully improves visibility, which outranks most formatting decisions.
Does an ATS reject my resume automatically?
In the overwhelming majority of cases, no. Enhancv put the question to 25 US recruiters working across healthcare, tech, construction, CPG, education, and retail, at organisations ranging from 120 to more than 50,000 employees, using over ten platforms including Workday, iCIMS, Greenhouse, and Bullhorn.
Twenty-three of the 25 said their systems do not auto-reject on formatting, design, or content. The 8% who had any content-based auto-rejection switched on were usually enforcing a genuine requirement rather than a layout rule.
Enhancv co-founder Volen Vulkov summarised the finding as software ranks and sorts, but humans decide. Recruiter Jan Tegze has reported that 90 to 95% or more of applications are seen by a person.
What does run automatically is the knockout question. If a posting requires work authorisation, a licence, or a specific certification and you answer no, you can be screened out without human review. That is a form field, not your resume.
Where did the 75% resume rejection statistic come from?
A sales pitch. Multiple investigations trace the figure to Preptel, a resume-optimisation vendor, in 2012. The company went out of business the following year. No survey was published, no sample size, no method.
Career consultant Christine Assaf investigated the claim and concluded it was created without any underlying study or context. It has since been repeated so widely that most job seekers assume it is established fact.
Enhancv asked recruiters where they had first encountered it. 68% said from job seekers on social media, 20% blamed career coaches and resume services recycling old advice, and 12% pointed at media coverage that never cited data.
Los Angeles recruiter Reggie Martin called the beat-the-bot genre a false narrative that takes advantage of job seekers. The practical cost is not just anxiety. It is that people strip the formatting and the substance that actually persuade a human reader.
What does an applicant tracking system actually do to your resume?
Four things, in order, and only the last one involves judgement.
It extracts. Your file is converted to raw text, which is why an image-based PDF with no text layer arrives as nothing at all.
It segments. Blocks of that text are classified as contact, experience, education, or skills, based on the section headings you used. Unconventional headings break this step.
It parses fields. Each block is broken into job title, company, start date, end date, description. This is where most silent damage happens.
Then it ranks. Your parsed record is scored against a search or a requisition, and you take a position in a sorted list. A recruiter works down that list until they have a shortlist.
Notice that rejection is not on the list. What feels like rejection is usually position 180 of 300 in a queue nobody had time to reach.
What is the difference between parsing, ranking, and rejection?
Conflating these three is the single most expensive mistake in ATS advice, because each one fails differently and each one has a different fix.
| Stage | What it means | How it fails | The fix |
|---|---|---|---|
| Parsing | Software reads your file into fields | Data lands in the wrong field or disappears | Single column, standard headings, real text |
| Searchability | Recruiter runs a keyword search | You never appear in their result set | The employer’s vocabulary, not your internal terms |
| Ranking | You are sorted against the requisition | You are present but buried | Relevance concentrated near the top of page one |
| Human scan | A person reads for a few seconds | Fit is not visible fast enough | A headline and opening that name the target role |
| Knockout | Hard requirement on the form | Genuine automatic screen-out | Answer the application questions carefully |
If your problem is ranking and you spend a weekend changing fonts, nothing improves. If your problem is parsing and you add keywords, nothing improves either. Diagnosis first is the whole point of the five-stage job search diagnostic.
Can your resume fail before a recruiter ever sees it?
Yes, but through parsing accidents rather than judgement. This is the legitimate core inside the myth, and it deserves attention precisely because the myth has made people cynical about it.
Workday runs the ATS at roughly 39% of the Fortune 500 and strips headers and footers during extraction. A name, email, or phone number placed in a Word header is dropped from the parsed profile entirely.
Tables are the second trap. Parsers read table cells linearly and cannot reconstruct the row-and-column relationship, so a tidy two-column skills grid can arrive as scrambled text.
Dates are the third. When a parser cannot compute your years of experience because the date format is unusual, you rank below candidates whose dates parsed cleanly, even with identical experience.
Then there is the form itself. On Workday, the structured fields are the application, and pre-fill accuracy is imperfect. People scroll past a truncated job title or a blank certifications section and submit anyway. The recruiter searches the fields, not your PDF.
Test it in five minutes: save your resume as plain text and read the result. Whatever survives that conversion is roughly what the parser sees. Missing contact details mean it sat in a header. Scrambled skills mean a table.
Does an ATS score of 90 mean I will get an interview?
No. A match score measures overlap between your document and a posting’s language. It does not measure credibility, seniority, or whether your experience solves the problem the role exists to solve.
ResumeAdapter’s pipeline data through Q1 2026 shows a median fit score of 48 out of 100 for unoptimised resumes, with 51% scoring under 50. The useful reading of that number is positional. It ranks, it rarely rejects.
Third-party scoring tools add another layer of confusion, because they score you against a model that is not the employer’s model. A 90 on a consumer checker and a recruiter’s actual search are different measurements of different things.
Chasing the score has a real failure mode too: keyword stuffing that reads fine to software and badly to the person who eventually opens the file.
So what actually gets your resume filtered out?
Volume and relevance, in that order. Ashby’s analysis of more than 100 million applications found open roles now average over 300 applications per hire, roughly triple the 2021 figure, while hires per recruiter sit at about seven per quarter.
That is the real gate. A recruiter with a 300-application pile and a shortlist to build reads the top of the sorted list and stops. Nobody rejected the bottom two-thirds. Nobody reached them.
Within that constraint, the things that move you up are unglamorous: a title that matches the posting, requirements answered in the employer’s own words, relevance concentrated in the first third of page one, and applying early. On timing, 52% of recruiters said the first 48 to 72 hours matter, because shortlists often form while a posting is still open.
How much of your editing time each application deserves is a separate question, and the 5, 15, and 45-minute framework answers it.
Methodology
The recruiter-side figures come from Enhancv’s structured interviews with 25 US recruiters across 10+ ATS platforms, plus its April 2026 survey of 1,066 US job seekers. Volume data comes from Ashby’s May 2026 analysis of more than 100 million applications across 200,000 jobs. Scoring distribution comes from ResumeAdapter’s Q1 2026 pipeline analysis. Parser behaviour is drawn from vendor and platform documentation on Workday extraction.
Two caveats worth naming. A 25-person qualitative sample establishes a strong directional finding, not a population statistic, and several sources here are companies that sell resume tools, which gives them an interest in how the ATS story is told. The claims that survived into this article are the ones that cut against the seller’s commercial interest, or that multiple independent parties report the same way.
Frequently asked questions about ATS and your resume
Should I use a PDF or a Word document?
Both parse reliably in 2026 as long as the PDF has a real text layer. The format to avoid is a scanned or image-based PDF, and print-to-PDF files that lack extractable text.
Will a two-column layout get my resume rejected?
Not rejected, but it can be misread. Parsers read across columns linearly, which can merge unrelated content. Single column is the safe choice on enterprise systems.
Can an ATS detect that I used AI?
Applicant tracking systems are not AI detectors. Humans notice generic phrasing, and Greenhouse’s 2026 research found 91% of recruiters have encountered some form of candidate deception, but that is a person’s judgement rather than a software flag.
Does keyword stuffing or white text work?
No, and it is increasingly detected. It also fails the moment a human opens the document, which is the stage that decides whether you get called.
Do small companies use an ATS?
Often yes, because tracking software is now bundled into job boards and HR suites. If you applied through a portal rather than by email, assume parsing happened.
How do I know if my resume parsed correctly?
Save it as plain text and read what survives, or check the pre-filled fields in the application form before submitting. On Workday, those fields are what recruiters search.
Is it worth paying for an ATS scan tool?
The parsing check is useful and often free. The match score is worth much less, because it compares you against the tool’s model rather than the employer’s search.
If the parse is clean and it is still silent
Clean parsing gets you into the queue. Position in the queue is a content and positioning problem, and that is the part a hiring-side reader can diagnose quickly. Book a free consultation, or read through the services and pricing first.
About the author. Dustin Harvey is a Certified Master Resume Writer (CMRW, CARW, CPRW) with an MBA in HR Management and roughly a decade on the hiring side, where he worked inside applicant tracking systems daily and decided who reached the interview pile. He now builds career documents for professionals from early career through the C-suite at HotResumeJob. Background is in the about section, client outcomes in the reviews, common questions in the FAQ, and further writing on the blog.