How an ATS Actually Reads Your Resume

"The ATS rejected me" is usually the wrong diagnosis. Most applicant tracking systems do not auto-reject candidates on their own. What they do is turn your file into structured data, and recruiters then search, filter and sort that data. If your resume parses badly, or uses different words than the job description, you simply never appear in the list a recruiter looks at. The fix is mechanical, which is good news: it is fully in your control.

Step 1: parsing — turning a file into fields

When you upload a PDF or DOCX, the ATS extracts the text and tries to map it into fields: name, contact details, job titles, employers, dates, education and skills. Workday, Greenhouse, Lever, iCIMS and Taleo all do some version of this. The parser reads text in order, top to bottom and left to right. Anything that breaks that order — or that is not text at all — is at risk.

Formatting that commonly breaks parsing

• Two-column layouts: many parsers read straight across both columns, so your skills list gets spliced into your job history.
• Text inside images, icons or charts: logos, skill-level bars and headshots are invisible to the parser.
• Tables and text boxes: content inside them is often skipped or pulled out of order.
• Contact details in the document header or footer: some parsers ignore those regions entirely.
• Creative section names: "Where I've Been" instead of "Experience" makes it harder to map your history.
• Unusual date formats: "Summer '23" is ambiguous; "Jun 2023 – Aug 2023" is not.

Step 2: matching — how your resume is compared to the job

Once your resume is structured, recruiters search it. A recruiter hiring a data engineer might search for "Airflow", "dbt" and "Snowflake", or filter for a title. Some systems also score candidates against the job description. Either way, the words matter. If the posting says "Amazon Web Services" and your resume only says "AWS", a strict search can miss you. Write the acronym and the full name once: "Amazon Web Services (AWS)".

Matching is about skills the job actually asks for, not stuffing. A resume that lists forty tools in white text, or repeats a keyword ten times, may score well in a machine check but fails the moment a human reads it — and a human always reads it before an interview.

Step 3: the recruiter's screen

Recruiters typically spend well under a minute on a first pass. They look at your most recent title, your employers, dates, and whether the top third of the page matches the role. A clean, single-column resume with a clear title line and strong first bullets survives this step far more often than a dense or decorative one.

A pre-submission checklist

1. Use one column, standard headings (Summary, Experience, Skills, Education) and a common font.
2. Put your name, email, phone, city and LinkedIn in the body of the page, not the header.
3. Save as PDF unless the posting asks for DOCX, and check the PDF text is selectable (not a scanned image).
4. Mirror the job's exact wording for skills you genuinely have, including both acronym and full name.
5. Lead each experience bullet with the result and the tool: "Cut build time 40% by moving CI to GitHub Actions."
6. Run the resume and the job description through an ATS checker and fix every missing keyword you can honestly claim.

You can paste any posting and your resume into the free STEMJobs ATS checker to see your match score, the missing keywords and formatting problems before you apply.