ATS friendly resume format: what actually breaks the parser

Resume formatting advice online is mostly superstition. Never use a PDF. Never use colour. Only Arial. One page or you are done. Most of it is repeated because it was repeated, not because anyone checked.

A short list of things genuinely break resume parsers. Everything else is a matter of taste. Here is the difference, plus a two minute test you can run right now on your own file.

What a parser is actually doing

When you upload a resume, the system does not look at it. It extracts the raw text and then tries to work out which chunk of that text is your name, which is your work history, which is your education, and where each job starts and ends.

It does that using position on the page, ordering, and recognisable heading words. Which means two things follow directly:

• If the extracted text comes out in the wrong order, everything downstream is wrong. Your job title lands in the company field and your dates end up attached to the wrong role.
• Whatever it extracted is what the recruiter searches against. Not your PDF. The recruiter may never open your file at all.

Five things that genuinely break it

1. Two column layouts

The biggest one, and the most common on templates sold as modern and professional. Some parsers handle columns correctly. Others read straight across the full width of the page, which interleaves your left column into your right and produces lines like Skills Software Engineer Intern Python June 2025.

You have no way of knowing which system a given company uses. A single column is the only layout that is safe everywhere, and it costs you nothing but a slightly longer document.

2. Tables used for alignment

Invisible tables are a common trick for lining up dates on the right. Depending on the parser, table cells get read in an order you did not intend, or the table structure gets dropped and the contents run together.

Use a right tab stop instead. Visually identical, structurally plain text, works everywhere.

3. Text inside document headers or footers

Word and Google Docs headers are a genuinely separate region of the file, and plenty of parsers skip them. People put their name, phone, and email up there because it looks tidy. The result is a resume that parses with no contact details at all.

Put your contact block in the body of the document. First lines of page one.

4. Icons standing in for words

An icon is a shape, not text. A phone icon next to a written phone number is harmless. A briefcase icon used instead of the word Experience means the parser has no heading to anchor on and does not know where your work history begins.

Keep the text labels. Add icons alongside them if you like the look.

5. Text baked into an image

The total failure case. A resume exported as an image, scanned, or designed in a tool that flattened the text produces a file with zero extractable characters. The parser gets nothing and you appear in no searches ever.

The select and copy test below catches this in about five seconds.

Six things everyone worries about that are fine

• PDFs. Text based PDFs parse correctly in every mainstream system. Send one unless the posting asks for DOCX.
• Two pages. Length is a human preference. One page is good discipline for students because it forces cuts, but nothing rejects page two.
• Colour. A coloured heading or a rule line is fine, and it helps a human scan. Just keep body text dark on light.
• Fonts. Any standard sans or serif. Nobody is filtering on Calibri versus Arial. Avoid anything decorative or very light.
• Bullet characters. Standard round bullets are fine. Avoid exotic glyphs that may not map to a character.
• A skills section. Fine and useful. Keep it plain comma separated text. Skip skill bars and star ratings, which convey nothing and parse as nothing.

The two minute test

1. 1

   Open your resume PDF

   In any viewer. Preview, Acrobat, or the browser.
2. 2

   Select all, then copy

   Ctrl+A then Ctrl+C, or Cmd on a Mac. If nothing highlights, your text is an image and you have found your problem already.
3. 3

   Paste into a plain text editor

   Notepad, TextEdit in plain text mode, or any code editor. Not Word, which will helpfully restore formatting and hide the issue.
4. 4

   Read what came out

   Are the sections in the order you wrote them? Is your contact block present? Did any two lines merge into one? Are dates still attached to the right jobs? If yes to all four, your format is fine and the problem is elsewhere.

The second test is free and more realistic. Start a Workday application at any large company and upload your resume. The autofill screen shows you the parsed fields directly. If your titles are in the company boxes and your dates are scrambled, that is the recruiter's view of you.

The safe structure

Boring, and it works in every system. Top to bottom:

1. Contact block. In the body, not the header. Name, email, phone, city, LinkedIn, GitHub or portfolio. Written out as text, not hidden behind icons.
2. Optional two line summary. Only if you will rewrite it per role. Otherwise skip it and use the space.
3. Experience. Reverse chronological. Each entry: job title, company, location, dates on one line, then bullets.
4. Projects. For students this can sit above Experience if it is stronger. Same structure as a job.
5. Education. School, degree and field, graduation month and year, and relevant coursework named specifically if you have room.
6. Skills. Plain comma separated text, grouped loosely by type.

Use the expected heading words. Experience, Education, Projects, Skills. Not:

• My Journey
• What I Bring
• Where I Have Been

Parsers anchor on the standard words. A creative heading is a small aesthetic win and a real parsing risk, and it is the easiest trade to decline.

One more thing that costs nothing. Name the file Firstname-Lastname-Resume.pdf. Not resume_final_v7.pdf. Recruiters download hundreds of these into one folder, and a file named after you is the one that gets found again.

A note on Workday specifically

Workday deserves its own paragraph because it is the system most large employers run, and it is the one everyone complains about.

Its particular habit is asking you to upload your resume and then re enter the same work history and education into individual form fields. It attempts to prefill those from your upload, and it does so imperfectly enough that you have to check every field regardless.

Two things follow. First, a clean single column resume with standard headings makes that prefill far more accurate, which saves you real minutes per application. Second, whatever is in those fields is what gets stored and searched, not your PDF. If the prefill is wrong and you leave it wrong, you are being filtered on garbage.

Check the fields. Every time. It is the least fun advice in this article and it is real.

If your format tests clean and you are still hearing nothing, the problem is upstream. Start with the five reasons resumes get rejected to find which one is actually yours, then work on the terms recruiters actually search for.