How to Write an ATS Resume

An ATS does not look at your resume the way you do. It sees text, not design. It extracts content into structured fields and scores how well that content matches the posting.

Two jobs happen in most systems. Parsing turns your file into text and maps it to the right fields. Ranking matches your text against keywords from the job ad. Some systems only rank. Others filter candidates out before a recruiter ever looks.

The stakes are concrete. Jobscan's 2025 survey of 384 recruiters found that 99.7% use keyword filters inside their ATS. Pika's analysis of 50,022 real resume scans found the average keyword match score was 65 out of 100. Half of all resumes scored 61 or less.

None of this means a machine decides your fate. Most systems hand a ranked list to a person. Your job is to rank high enough to be on that list.

The most common reason a resume underperforms is not formatting. It is keyword mismatch. Jobscan's analysis is blunt: a candidate with ten years of exactly the right experience can score below a candidate with five years of less relevant experience, if the second candidate matched more of the posting's phrasing.

Formatting still matters, because a layout the parser cannot read loses your content entirely. But once the file parses cleanly, relevance decides whether you make the shortlist.

You rarely know which ATS a company uses, so build for the least capable one you might face. A plain single column with standard headings passes all of them. Fancy layouts pass only the ones built for them. Start with our ATS resume templates, which follow every rule below.

Every section below uses a heading name a parser already knows. That is the whole trick: the section names do the mapping for you.

Keywords are the literal words a recruiter or an ATS searches for, and matching is often exact. Your job is to mirror the phrasing of the posting, not to write synonyms.

Pull the hard skills, tools, and the job title from the job description, and place them where they belong: in the skills section, inside real experience bullets, and in your summary. Use the posting's exact terms. If it says REST APIs, write REST APIs, not web services.

There is no magic keyword count. Relevance and natural placement beat volume. Stuffing a resume with terms you have never used is easy for a recruiter to spot, and it actively hurts you in modern systems.

Never hide keywords in white text or zero-size font. Parsers read hidden text, and any recruiter who pastes your resume into plain text sees the block instantly. It reads as dishonest and can get you rejected.

A bullet that carries a keyword and a result serves both audiences. The machine gets the exact term. The recruiter gets a number and a named tool. Aim to write every important bullet that way.

Use a strong action verb, a specific result, and a named tool or metric. Start each bullet with the action. Put the number in the same sentence as the tool.

Modern systems parse text-based PDFs reliably. The failure mode is an image-based PDF, where the characters are pixels, not text. Word, Google Docs, and most resume builders keep a real text layer by default.

If a posting does not specify a format, .docx is the safest default. Design-tool exports, like anything built in Canva or InDesign, use embedded text boxes and custom fonts that break parsing. Save those for portfolios.

Before you submit, run the copy-paste test. Paste your resume into a plain text editor. If the text comes out in the right order with your headings and keywords intact, the file will parse. If it comes out garbled or empty, fix the file first.

One page is the standard for early-career candidates. For candidates with five or more years of experience, two pages is acceptable and often better. Cutting real accomplishments to force one page removes the detail and metrics that both the ATS and the recruiter want.

A lot of what circulates about ATS resumes is outdated or plain wrong. Here is the short version.