The five ATS formatting mistakes that quietly kill your resume
Each of these looks completely normal on screen and falls apart the moment a parser reads it. Here is the symptom, the cause, and the one-minute fix for each.
Nearly every applicant tracking system filters candidates before a person ever opens the file. If the parser can't extract your dates, your contact details, or your skills cleanly, the recruiter is working from a broken record — and you never find out why you went quiet.
The five mistakes below account for most of that damage. Each one looks fine on screen, which is exactly why they survive so many rounds of proofreading.
1. Dates the parser can't line up
Recruiters commonly filter by years of experience, so if a parser can't read your dates, it can't calculate that number — and an unreadable "Work History" section often gets left blank rather than guessed at.
The symptom: your work history imports empty, or shows a role running from "January 1" to "Present" with no year attached.
The cause: inconsistent or abbreviated date formats.
| Breaks parsing | Parses cleanly |
|---|---|
Jan '21 – Mar '23 | Jan 2021 – Mar 2023 |
2021 – 2023 (no month) | January 2021 – March 2023 |
1/2021 – 3/2023 | 01/2021 – 03/2023 |
The fix: pick one format and use it everywhere on the document, including education and certifications. Apostrophed years and month-less ranges are the two most common culprits.
2. Fonts and icons that turn into noise
The symptom: your name renders as [NULL], or "Profile" comes out as
"Pro?le" with bullet points replaced by stray letters.
The cause: most parsers scrape text rather than render the page visually. If a font isn't embedded or widely supported, or if you've swapped a word for an icon (a phone glyph instead of "Phone:"), the scraper has no mapping for that shape and substitutes garbage or nothing at all.
How to check it yourself: copy the full text of your resume out of the
PDF and paste it into a plain .txt file. Whatever comes out is close to
what the ATS actually sees. If it reads like static, the font is the problem.
The fix: stick to fonts every system ships with — Arial, Calibri, Helvetica, Georgia, Garamond, Tahoma, Verdana. Replace decorative icons with the plain word they stand in for.
3. Information hidden in headers, footers, and text boxes
The symptom: the resume reads well to a person, but the recruiter says they can't find your contact information or your skills section.
The cause: headers, footers, and floating text boxes live on a separate layer of the document. Many parsers only read the main body and skip that layer entirely — so anything placed there effectively doesn't exist to the system.
A quick way to check: select all the text in your document (Ctrl+A or Cmd+A). If content in your header or a sidebar doesn't highlight along with everything else, a parser likely can't see it either.
The fix: move contact details and any load-bearing information into the main body of the document. Use bold text and size, not a text box, to create visual hierarchy.
4. Section headings the system doesn't recognize
The symptom: your work experience gets miscategorized under education, or your professional summary disappears from the parsed output entirely.
The cause: ATS software looks for a fixed set of expected headings to sort your content into fields. Creative labels like "My Journey," "Where I've Been," or "The Toolkit" don't match anything in that list, so the content underneath lands in an unsorted blob — or gets dropped.
The fix: use the standard headings recruiters and parsers both expect: "Work Experience" (or "Professional Experience"), "Education," and "Skills." This is one place where a little creativity actively costs you.
5. Columns, tables, and graphics that scramble the reading order
The symptom: the parsed text reads out of order — a line from your left column merges mid-sentence with a line from the right, like a shuffled deck of cards.
The cause: a table or multi-column layout has a visual order that's obvious to a human eye but not to a parser, which typically reads left to right, top to bottom, one line at a time regardless of columns. A skill bar graphic showing "80% Java" also extracts as nothing at all — there's no text there to read.
The fix: use a single-column layout throughout, and describe skill level in words instead of a chart — "Java (Expert)" rather than a bar graph.
Checking your own resume
Before you send it anywhere, copy the entire document and paste it into a plain text editor. That output is roughly what the system will see. Anything scrambled, duplicated, or missing there is worth fixing before you apply again — it takes a few minutes and it's the difference between your resume being read and being silently discarded.
