The "75% rejected" stat is everywhere. It's also unsourced.

01The "75%" myth, traced

Search "ATS rejection rate" and you'll get the same figure back almost everywhere: 75% of resumes never reach a human. It's quoted with total confidence and almost never with a citation attached.

That's because there isn't one to attach. No peer-reviewed study, published methodology, or named sample size backs this number. It appears to have originated as a marketing claim from a resume-optimization vendor years ago, then spread through career blogs the way unsourced statistics do — repeated often enough that repetition became the evidence.

A statistic that nobody can trace to a primary source isn't a statistic. It's a rumor with a percentage sign.

02What's actually documented

The most credible research on this problem isn't from a resume tool — it's from Harvard Business School and Accenture's "Hidden Workers: Untapped Talent" report (2021), based on a survey of employers across the US, UK, and Germany.

That's a meaningfully different claim than "75% get rejected." It says the filtering problem is real and widespread — but it's a byproduct of overly rigid configuration (exact-match keyword rules, rigid experience cutoffs, employment-gap penalties), not a fixed rejection rate baked into every ATS everywhere.

03The human side of the funnel

Passing the ATS isn't the finish line — it just gets you into a much smaller, much faster pile that a human still has to look at.

The practical takeaway: a resume that's technically parseable but generic still loses at the human-scan stage. Passing the ATS earns you the 7 seconds. What you do with those 7 seconds — a clear title match, a scannable summary, numbers in the first few bullets — decides the rest.

04Formatting failures vs. content failures

Most resume advice fixates on formatting — no tables, no columns, no icons. That advice is correct, but it's solving the smaller half of the problem. Our own full ATS guide breaks down exactly how parsing works and where formatting-driven failures happen (text boxes, sidebars, unparseable headers).

The bigger, less-discussed failure mode is content: a resume with no formatting issues at all can still rank poorly if it doesn't contain the tools, titles, and phrases the job description actually asks for. Formatting gets a resume read correctly. Content gets it ranked highly. Both matter; only one of them gets talked about.

05The one thing the data agrees on

Across every source in this article — vendor research, academic studies, recruiter-reported behavior — one finding shows up consistently: a resume tailored to the specific job description outperforms a generic one sent to every posting.

Matching the exact job title, mirroring the tools and certifications named in the posting, and reflecting the language of the role — not just the industry — is the most consistently supported lever in all of this research. It's also the one most job seekers skip, because doing it by hand for every application takes real time.

06Sources

Last reviewed 2026. If you find a primary source for the "75% rejected" figure that includes a methodology and sample size, we'll update this page and credit it.

07Frequently asked questions

What percentage of resumes are rejected by ATS?

There is no verifiable study behind the commonly repeated "75%" figure. What's documented: 88% of employers surveyed by Harvard Business School and Accenture say qualified candidates get filtered out — but no credible source has produced a single defensible "rejection rate."

Does a resume need to be perfect to pass an ATS?

No. Most ATS platforms rank and sort applications rather than auto-rejecting them outright. The goal is ranking high enough — and being parsed cleanly enough — that a recruiter opens your file at all.

What matters more: formatting or content?

Content. Formatting failures are real but fixable in minutes. Keyword and relevance gaps between your resume and the job description are the more common, more consequential problem.

How much time does a recruiter spend reading a resume?

TheLadders' eye-tracking research found roughly 6-7 seconds on the initial scan. It's an older study, but directionally consistent with what recruiters describe today.

Does tailoring a resume to each job actually help?

Yes — it's the most consistently supported finding across this research. Matching the job title, tools, and language of a specific posting measurably improves how a resume ranks.