The myth vs. the reality

Search "does ATS reject resumes" and you'll find a lot of confident, specific-sounding claims: a fixed percentage of resumes get rejected before a human sees them, systems auto-reject anything under a certain keyword match score, formatting mistakes guarantee automatic disqualification. Some of this is exaggerated, some is outdated, and some varies so much company to company that a universal claim about "the ATS" is misleading on its face — there isn't one ATS. There are dozens of different systems (Workday, Greenhouse, Lever, iCIMS, and many smaller ones), each configured differently by each company that uses it, and they don't all behave the same way.

What's true across most of them is more mundane than the myths suggest, and understanding the mundane, real version is actually more useful than the dramatic one.

What an ATS actually does

At its core, an applicant tracking system is a database with a form attached. When you apply to a job through a company's careers page, three things typically happen:

  1. Your resume file gets stored, usually alongside the text extracted from it.
  2. Some of that extracted text gets mapped into structured fields — name, email, phone, sometimes work history and education — often to auto-fill the application form so you don't have to retype everything by hand.
  3. The system becomes searchable by the recruiter or hiring manager, who can filter or search applicants by keyword, requirement, or other criteria they set up.

That third point is the one that matters most and gets the least attention. In a lot of real-world hiring, especially at larger companies, a recruiter isn't waiting for an algorithm to rank-order every applicant automatically — they're searching and filtering a pool of candidates themselves, the same way you'd search a spreadsheet. Which means the practical question isn't "will an algorithm mysteriously reject me," it's "if a recruiter searches for the terms in this job posting, does my resume come up, and does it read clearly once it does."

Where automated ranking or filtering does happen

Some systems do include scoring or ranking features, and some companies do configure hard filters — a minimum years-of-experience field, a required certification, a location radius — that can screen out applications automatically before a human ever opens them. This is real, but it's usually based on structured fields the applicant fills in directly on the form (a dropdown for "years of experience," a checkbox for "authorized to work in X country"), not a mysterious algorithmic read of your resume's prose. Answering those form fields honestly and completely matters as much as the resume content itself, sometimes more, since a hard filter based on a form answer bypasses your resume entirely.

What breaks a resume in this pipeline

Given that structure, the real risks aren't mystical — they're the same practical formatting issues covered in more detail elsewhere: two-column layouts that scramble text order during extraction, information hidden inside header/footer fields the extraction step skips, text embedded in images instead of typed out, and non-standard section headings that make it harder for a human doing a keyword search to quickly confirm you have the experience they're looking for. None of this is a system "rejecting" you outright — it's your information failing to come through clearly to whoever, or whatever, reads it next.

What doesn't actually matter as much as job seekers fear

  • A slightly unusual font. As long as it's a standard, readable typeface (not a script or decorative font), font choice has minimal effect on parsing or scoring.
  • Using synonyms instead of a job posting's exact phrase, most of the time. A human recruiter searching a database will often try more than one term, and a reasonably close synonym ("customer support" vs. "customer service") is not typically the difference between being found and not — though matching the posting's exact language where it's honestly accurate doesn't hurt.
  • A resume that's "only" a 70% keyword match on some third-party scoring tool. These match scores (including the one on Craftume's own scanner) are useful directional signals for what to strengthen, not a pass/fail gate that mirrors any specific employer's actual system.

What actually helps

  1. Include the real keywords from the job posting, in context, where they're honestly true of your background. Not stuffed into a hidden block of text, but used naturally in your skills section and bullet points.
  2. Use standard section headings and a clean, single-column format, so whatever extracts your resume gets an accurate, ordered read of it.
  3. Fill out application form fields completely and honestly, since those structured fields are more likely than resume prose to feed an automated filter.
  4. Write for the human who eventually searches or reads the results, not for an imagined algorithm — because in most real hiring pipelines, a person is still the one making the actual decision, just often after a search or filter step narrows the pool first.

If you want a concrete, practical check rather than general theory, running your resume and a specific job description through Craftume's ATS Resume Scanner shows you the real overlap in language between the two — a useful proxy for "would a recruiter searching this posting's key terms find my resume," even though it's not a simulation of any one company's exact system.

FAQ

Do all companies use an ATS? Most mid-size and large companies do, in some form, if only as a database and application-tracking tool rather than an automated screener. Small businesses and startups sometimes review applications directly via email or a simple form instead.

Can formatting alone get a resume automatically rejected? Severe formatting issues (a two-column layout that scrambles your name and contact info, or text buried in an image) can cause your information to come through wrong or incomplete, which functions like a rejection in practice even without an explicit "reject" decision being made. It's a data quality problem more than a judgment call by the system.

Should I apply the exact keywords from a job posting even if I'd phrase it differently myself? Where the posting's term is honestly accurate to your experience, yes — it helps both a keyword search and a fast human skim. Don't claim a specific keyword (a tool, a certification, a methodology) you haven't actually used, since that gap becomes obvious the moment it's asked about directly.

Is it true that a human never reads a resume until it passes an ATS score? It varies enormously by company and role. Some postings get reviewed by a human quickly regardless of any score; others go through a higher-volume automated filtering step first. There's no single universal pipeline, which is exactly why formatting for clean, honest, searchable content is a safer bet than trying to game one specific hypothesized algorithm.

Does having a personal website or portfolio link hurt ATS parsing? No — a plain text link in your contact section is simple text like any other and doesn't cause parsing problems on its own.