What an ATS actually does with your resume
When you submit, the system extracts the raw text and attempts to structure it. It looks for headings it recognises, date ranges it can parse, and job titles it can attach to employers.
If it succeeds, your experience lands in the right fields and you appear in the recruiter's searches. If it fails, your resume is still there — it is just poorly indexed, so you do not surface when someone searches for the skill you actually have.
An ATS is a filing system with a search box, not a judge. In most configurations it does not score you, does not reject you, and does not decide anything. A human runs a search and reads what comes back. Being ATS-friendly means being findable, not being approved.
What genuinely breaks parsing
In rough order of how often it causes real damage.
1. Multi-column layouts
This is the big one. Text extraction generally reads left to right across the full page width. A two-column design means a line from your sidebar gets interleaved with a line from your main column, and the result is scrambled — job titles fused to unrelated skills, dates attached to the wrong employer.
Those elegant templates with a shaded skills rail down the side are the single most common cause of a mangled parse.
2. Text inside images or graphics
A skills chart, a logo containing your name, a header saved as a graphic — none of it exists as far as the parser is concerned. Text that is not text is invisible, and no amount of keyword optimisation recovers it.
3. Headers and footers
Contact details placed in the document header are frequently dropped, because header and footer regions are often excluded from extraction. A resume that parses perfectly but loses your phone number and email is worse than one with a scrambled skills section.
Put contact details in the body of the document, at the top.
4. Non-standard section headings
Parsers look for headings they recognise. "Work Experience", "Education", "Skills" are recognised. "Where I've Made an Impact" and "My Journey" are not, and the content underneath them may never be categorised.
This is a place where creativity costs you and conventionality is free.
5. Tables for layout
Tables used to position content produce the same interleaving problem as columns. A table used for genuinely tabular data is usually fine; a table used as an invisible layout grid is not.
| Formatting choice | Effect on parsing |
|---|---|
| Two-column layout | Breaks it — reading order scrambles |
| Text inside an image | Invisible entirely |
| Contact details in the page header | Frequently dropped |
| Creative section headings | Content may go uncategorised |
| Layout tables | Interleaves unrelated lines |
| Standard fonts | No effect — use any readable font |
| Bold and italic | No effect |
| Two pages instead of one | No effect on parsing |
| .docx or PDF | Both fine in modern systems |
Myths worth dropping
- "Never use a PDF." Modern systems parse text-based PDFs fine. A PDF exported from a design tool as flattened artwork is the actual problem — that is an image, not a PDF issue.
- "You must use Arial or Times New Roman." Font choice has no bearing on text extraction. Any embedded, readable font works.
- "Keep it to one page or the ATS will reject it." Nothing in the parser counts pages.
- "Hide white keywords in the background to beat the filter." This is extracted as visible text, is obvious to any recruiter who reads it, and reliably ends the application.
- "The ATS rejects you automatically below a score." Most configurations do not score at all. A recruiter searches and reads.
See how your resume reads against a real posting
Upload once and check your experience against a specific job description before you spend an hour on the application.
Check a matchHow to test your own resume in two minutes
You do not need a tool for the basic check. You need a plain text file.
- Open your resume and select all of it, then copy.
- Paste into a plain text editor — Notepad, TextEdit in plain mode, or any code editor.
- Read what came out. This is roughly what the parser sees.
- Check the order is still sensible, that no lines are interleaved, and that your name, phone and email are all present.
- Check each job still sits with its own dates and employer.
If the plain text version reads correctly top to bottom, your formatting is fine. If it is scrambled, the layout is the cause, and no keyword adjustment will fix it.
What to do about keywords
Keywords matter, but not as a trick. Recruiters search the database for the skills they need, so if a posting asks for a technology you have used, that word should appear in the context of the job where you used it.
Written honestly, that is just an accurate resume. Written as a keyword list stapled to the bottom, it surfaces you for searches you cannot back up in an interview, which wastes everyone's time including yours.
Being findable for work you have actually done is the whole goal. Anything beyond that is optimising for the wrong reader.