What the ATS Actually Sees on Your Resume (with examples)

If you’ve ever submitted a beautiful, carefully-designed resume to a job portal and heard nothing back, this article is the answer. The recruiter never saw it the way you designed it. They saw what the applicant tracking system extracted, which is usually not the same document at all.

Below: two example resumes, what the parser extracts from each, and what that means for whether a human ever sees your application.

The 30-second version

An ATS reads your resume top to bottom, left to right, treating it as structured text. Anything that breaks that linear reading order — multi- column layouts, tables, text boxes, images with text inside them, content in the docx header zone — gets scrambled, dropped, or misplaced. The recruiter’s database fills with garbage in your fields, and your application is automatically ranked low for the job’s required keywords.

Example 1 — the fancy two-column template

Here’s a typical “modern” resume template you’d find on Canva or in Word’s default gallery. Single page, two columns, photo in a circle in the top left, sidebar with skills and contact info, main column with experience.

Looks great on a screen. Now here’s what a standard ATS extracts from it:

JANE DOEEMAIL: jane@   SKILLS PROJECTS LANG
PHONE: 555-EXPER       Pyth React Java AWS
IENCE Senior Engineer  Docker K8s SQL React
Acme Corp 2022-PresentL ed migration of pay
ments stack Reduced    deploy time from 18
to 4 min Built rate-   limiter handling 8k
rps EDUCATION BSc CS   MIT 2018 GPA 3.8

That’s the literal parsed output. Notice what happened:

The name “Jane Doe” gets mashed against “EMAIL:” because the parser read row-by-row across both columns
Skills are interleaved with experience — the right-hand sidebar got read alongside the left-hand main content
“Engineer” appears inside the middle of an unrelated word because the layout broke parser tokenization
The contact email is missing entirely — it was in the docx header zone, which this parser silently skipped

That document is unrecoverable. A recruiter searching their ATS for “Kubernetes” or “K8s” might find Jane’s resume (the keywords are there somewhere), but a recruiter doing the more common F-pattern visual scan sees gibberish and moves on. A hiring manager filtering for “5+ years distributed systems” probably won’t even see her in the search results because the parser couldn’t extract her job title cleanly.

Example 2 — the same person, single-column template

Same Jane, same career, same numbers — just rebuilt in a single-column, ATS-safe format like the one in our free template bundle. Here’s what the parser extracts:

NAME: Jane Doe
CONTACT: jane@email.com | 555-0123
LOCATION: Brooklyn, NY

PROFESSIONAL SUMMARY
Senior backend engineer with 7 years building payments systems.
Led $400M GMV infrastructure at Acme. Skilled in Python, AWS,
Kubernetes, distributed systems.

EXPERIENCE
  Senior Engineer @ Acme Corp
  2022 - Present
  - Led payments stack migration
  - Cut deploy time 18m → 4m
  - Built rate limiter, 8k rps

EDUCATION
  BSc CS, MIT, 2018, GPA 3.8

SKILLS
  Python, React, AWS, Docker, K8s, SQL,
  PostgreSQL, Kafka, distributed systems

Every field landed where it belongs. The recruiter’s database now correctly shows:

Name field: Jane Doe
Email field: jane@email.com
Most recent title: Senior Engineer
Most recent company: Acme Corp
Skills: Python, React, AWS, Docker, K8s, SQL, PostgreSQL, Kafka, distributed systems

When the recruiter searches “Kubernetes engineer” inside Workday, Jane shows up at the top of the candidate list. When the F-pattern visual scan happens, the top-left quadrant tells the whole story in five seconds.

Same person. Same skills. Wildly different outcome.

Why parsers break on “fancy” layouts

ATS parsers were designed in the early 2010s when most resumes were single- column Word documents. They’ve improved since, but they still rely on the underlying assumption that text flows linearly. When you add structural complexity, the parser has to guess at the right reading order.

Common parser-breakers, ranked by how often they cause damage:

Resume element	How parsers handle it	Severity
Two- or three-column layouts	Reads row-by-row across all columns	Catastrophic
Tables (even invisible ones)	Flattens into single line or skips	High
Text boxes	Often skipped entirely	High
Headers/footers	Many ATSes ignore the docx header zone	High — contact info loss
Custom font (not installed on parser)	Falls back to system font, sometimes mangles glyphs	Medium
Drawing-character bullets (▸, ★, ✓)	Replaced with `?` or dropped	Low — cosmetic
Images containing text	Not OCR’d by most ATSes	Severity depends on what’s in image
Multi-page (>2 pages)	Sometimes only first page parsed	High for long resumes

The single safe assumption is: structure = enemy of parsing. Strip every visual flourish that isn’t required for human readability.

How to test what an ATS sees on your resume

You don’t need ATS-vendor software. There’s a free test that takes 30 seconds and is surprisingly accurate.

The copy-paste test

Open your .docx resume in Word, Pages, or Google Docs
Press Ctrl-A (Windows) or Cmd-A (Mac) to select all
Copy the selection
Paste into a plain-text editor — Notepad, TextEdit in plain-text mode, or VS Code with a new untitled file

What you see in the plain-text editor is approximately what the ATS extracts.

If the result is clean, sensible, in the right order — you’re fine. If it’s scrambled, has fields out of order, has random whitespace, or is missing your contact info — your resume needs to be rebuilt before you submit it anywhere else.

What “clean” looks like

First Lastname
city, state | phone | email

PROFESSIONAL SUMMARY
[summary text...]

EXPERIENCE
[role 1...]
[role 2...]
...

In that order. No interleaving. No missing fields.

What “broken” looks like

Skills section content appears between job entries
Email or phone is missing
Job titles are merged with company names with no separator
“Education” header appears before all your experience
Random text appears mid-sentence

What to do if your resume fails the test

Two options:

Identify what’s breaking it — usually columns or tables. Strip those, reformat into a single column, and re-test.
Start from a known-clean template. Faster and lower risk than trying to debug a complex layout. Our free ATS bundle gives you a parser-safe starting point.

Either way: don’t submit another application until your resume passes the copy-paste test.

Frequently asked questions

Don’t modern ATSes handle two-column resumes now?

Some do, some don’t. The ones used by the Fortune 500 (Workday, SuccessFactors, iCIMS) handle them with mixed reliability — some columns parse correctly, some don’t, depending on how the columns were created in the source file. The risk-adjusted answer is: don’t bet your job search on the assumption that a parser will get it right.

What about PDFs — do they parse better than .docx?

Generally no. PDFs are layout descriptions, not structured documents. ATSes have to reconstruct reading order from geometry, which fails on creative layouts and works fine on simple ones. Use .docx when the portal accepts it; export to PDF only when required.

My current resume has gotten me interviews before. Should I change it?

Maybe not. If a resume is converting at a rate you’re happy with, don’t break what’s working. But the copy-paste test takes 30 seconds — if it reveals a parsing disaster you weren’t aware of, you might be screening out opportunities you’d otherwise have gotten.

Will this template work outside the US?

The bundle includes notes for US, UK, and Canada. The single biggest regional variant is the photo — US/UK/CA resumes should not include a photo (it triggers bias-prevention rules and the resume gets discarded), while many European and Latin American markets expect one. Read the included notes before submitting internationally.

Free ATS resume template (.docx, 2026) — the bundle.
Single-column resume template for Word — why single column wins.
hireformat homepage — the live ATS comparison demo.