How to Write an ATS Resume 2026

ATS resume formatting mistakes cost good candidates interviews every day — and most of them have no idea it's happening.

I've spent a lot of time looking at why good candidates get filtered out before a human ever sees their resume. Most of the time it isn't talent. It's formatting choices and keyword gaps they didn't know were costing them. This post pulls together what actually moves the needle on ATS ranking in 2026, based on what we've consistently seen work for early-career job seekers, what recruiters say they actually do, and the research on how these systems parse and rank resumes.

Let me start by killing the myth that's been scaring people for over a decade.

Is the ATS Really Rejecting Your Resume Automatically?

You've probably heard that "75% of resumes get auto-rejected by the ATS." That number traces back to a 2012 sales pitch from a company that no longer exists. It has no methodology behind it. When we surveyed recruiters, the vast majority confirmed their ATS does not auto-reject based on content.

Applicant tracking systems like Workday, Greenhouse, Lever, and Taleo parse your resume into structured fields, then rank you against the job description using keyword matching. Recruiters review the top 10 to 20 candidates. Being ranked 150th out of 180 means nobody scrolls far enough to find you, which produces the same outcome as a rejection.

And the pile is enormous. Workday processed 173 million applications in the first half of 2024, up 31% year over year, while job openings grew only 7%. The problem isn't a robot saying no. It's volume, plus a resume that ranks too low to get seen. Everything below is about ranking higher.

How Should You Format a Resume So ATS Can Read It?

Before keywords matter, the system has to read your resume correctly. A surprising number of strong candidates lose here because they used a pretty template that scrambles when parsed. These are hard requirements, not style preferences.

• Use a single column. Multi-column and sidebar layouts are the single most common reason resumes parse wrong. The parser reads left to right across both columns at once, so your job titles end up mixed with your skills. One column, top to bottom.
• Use standard section headings. Write "Work Experience," "Education," "Skills," and "Summary." Creative headers like "My Journey" or "Tech Arsenal" don't get recognized, and the content underneath them can disappear from the parsed version entirely.
• Put contact info in the body, not the header or footer. Most ATS platforms skip headers and footers during parsing. Your name, email, and phone should be the first lines of actual body text.
• No tables, text boxes, images, or graphics. Anything inside these elements gets skipped or pulled out in the wrong order. That includes those skill bars and icon sets from design tools.
• Skip Canva and InDesign exports. Design-tool PDFs are full of text boxes and custom fonts that break parsing. Use a clean .docx instead.
• File format: PDF or doc.x. Never submit .pages, .jpg, or .png. If the posting doesn't specify, .docx or PDF is safest.
• Use standard fonts and round bullets. Arial, Calibri, Georgia, Helvetica, Times New Roman. Decorative fonts render as symbols or drop out. Fancy bullet characters often turn into garbage in the parsed text.

Here's the template I found with the highest interview rate after applying to 200+ companies with varying resume templates. (Landed me offers at Meta, Microsoft, Citadel, and more!)

One quick self-test: open your resume, select all, copy, and paste it into a plain text document. If the order is jumbled or chunks go missing, that's roughly what the parser sees. Fix it there before you fix anything else.

How Do You Find the Right Keywords for an ATS Resume?

Once you're parsable, ranking comes down to keywords. The mistake most people make is guessing what to include. Pull them straight from the source instead.

Here's the process I'd run for any role:

1. Find five real postings for your target role. Any term that shows up in three or more of them is definitional. It has to appear on your resume. This is also how you catch the exact phrasing employers use, which matters because exact matches still rank higher than semantic ones.
2. Cross-reference O*NET. It lists federally standardized skill terms for almost every role, and these are the kinds of terms ATS synonym tables tend to draw from.
3. Sort into three tiers. Tier 1 is the hard skills and tools listed as requirements, match these exactly. Tier 2 is the soft skills and competencies that repeat across the description. Tier 3 is company- and industry-specific language that signals fit.
4. Ask the recruiter if you get the chance. A simple "What are the must-have skills the hiring manager is screening on?" gives you the list directly.

Aim for 15 to 25 keywords total, with 5 to 8 pulled straight from the specific posting. A good target is matching 70 to 80% of the keywords in that job description.

Placement matters as much as presence. A keyword inside a work experience bullet ranks higher than the same word sitting in a skills list, because the system reads it as evidence you actually did the thing. So spread your terms out: 3 to 5 in the professional summary, more woven through your bullets, and the rest in a dedicated skills section.

One more thing the 2026 systems care about: don't dump 40 keywords in a comma-separated wall. Workday and Greenhouse flag unnatural keyword density and will downscore you for it. Use each key term two or three times in real context, not twenty times in a list.

A small detail that quietly doubles your coverage: spell out acronyms with both forms. Write "Search Engine Optimization (SEO)" and "Project Management (PM)." Different systems search for different versions, so including both covers you either way.

What Makes a Resume Bullet Actually Work in ATS and Human Review?

Getting ranked high gets you read. After that, a recruiter spends about six seconds on the first scan. Your bullets have to do two jobs at once: carry keywords for the ATS, and prove real experience to the human.

The formula is action verb + keyword in context + quantified result, with the tool named where it applies.

Here's the difference in practice.

Weak, and the kind of thing AI spits out by default:

"Spearheaded cross-functional initiatives to drive scalable outcomes and improve team efficiency."

Strong:

"Reduced order-processing time from 4 hours to 90 minutes by automating invoice matching in SAP, saving roughly $180K a year."

The second one has a keyword (SAP), a number, a real before-and-after, and something you could explain out loud. That last part matters more in 2026 than it used to, which I'll come back to.

For your skills section, list actual tools and platforms, not vague categories:

Technical Skills: Python, SQL, Tableau, AWS (EC2, Lambda), Docker Methodologies: Agile, Scrum, Six Sigma Tools: Jira, Confluence, Salesforce, Figma Certifications: PMP, AWS Solutions Architect – Associate

One note on what's rising and falling. "AI literacy" and AI-related skills are the fastest-growing keyword category across basically every industry right now, up about 30% year over year in job postings. For technical roles, "Prompt Engineering," "RAG," and "LLMs" are now legitimate resume terms. Meanwhile "Proficient in Microsoft Office" and "Problem-Solving" as standalone claims are dead weight you should cut.

Why Does Tailoring Every Application Give You a Real Edge?

This is where most candidates lose to people no more qualified than they are. About 54% of job seekers never tailor their resume, and that's your opening.

Tailoring doesn't mean rewriting everything. At minimum, swap in the Tier 1 keywords from each posting and mirror the exact job title. A resume that says "Data Analyst" for a Data Analyst req ranks better than one that says "Data Specialist," even if the work was identical. With application volume what it is, the generic resume gets buried under tailored competitors every time.

The candidates we've worked with who track their tailoring systematically — noting which keywords they swapped in for each role — tend to spot patterns in what's actually driving callbacks. Simplify's Job Tracker lets you do exactly that: attach the tailored version of your resume to each application and monitor which customizations lead to responses.

Should You Use AI to Write Your ATS Resume?

ATS platforms do not detect AI-generated text. Testing across eight major systems, including Workday, Greenhouse, and Lever, found ChatGPT-written content parses exactly the same as human-written content. Parsing failures come from formatting, not from who or what wrote the words.

Recruiters mostly aren't running AI detectors either. Surveys put it at around 5 to 8% of recruiters, and the tools themselves are unreliable anyway. GPTZero flagged formal professional writing as AI-generated at rates over 20%, sometimes higher, which means a perfectly human resume can get falsely flagged. One big-tech pilot got rolled back because it kept flagging non-native English writers.

The real filter is recruiter pattern recognition. In blind tests, recruiters spotted AI-written resumes about 83% of the time from a five-second skim. They catch it by the tells: generic strong-verb chains, every bullet the exact same length, platitude openings, and zero specific detail. That's a quality problem, not a detection problem.

The fix is the workflow. A large field study from NBER and MIT — nearly half a million job seekers — found that AI writing help increased hires by about 8% and wages by about 8%. But it worked as an editor on real human content, not as a generator. So:

1. Write your bullets yourself first, in rough form, focused on what you actually did.
2. Ask AI to sharpen, not invent. Prompt it with "Here are my bullets for this role, help me make them clearer and more impact-focused, keep all my specific details." Don't prompt "write me a resume for a senior engineer."
3. Edit it back to human. Vary bullet length, restore your real numbers and tool names, and delete any phrase you couldn't defend out loud in 60 seconds.
4. Run it through an ATS scanner against the target posting and close any genuine keyword gaps.

That last instinct — only keeping what you can defend — matters because the interview is now the consistency check. Recruiters increasingly ask you to walk through the exact context behind a bullet. A fabricated or AI-hallucinated metric falls apart fast there, and it follows you at that company. If your resume says you cut processing time to 90 minutes, be ready to explain how.

Dos and Don'ts

Do:

• Use a single column with standard section headings
• Put contact info in the body as the first lines of text
• Pull keywords from five real postings and mirror exact phrasing
• Put keywords inside bullets, with numbers and named tools
• Tailor the title and Tier 1 keywords to every posting
• Use AI to edit your own writing, then make it sound like you

Don't:

• Use two-column or sidebar templates from design tools
• Invent creative headers like "Tech Arsenal" or "My Journey"
• Dump 40 keywords into a comma-separated wall
• Submit .pages or image files
• Generate a whole resume from scratch with AI
• Put a number on your resume you can't explain in an interview

None of this is about gaming a machine. It's about not getting filtered out for reasons that have nothing to do with whether you can do the job. Get the formatting right so you're readable, match the language so you rank, and keep the specifics so a human believes you.

When you're ready to apply, Simplify keeps the whole process together — from parsing checks to tracking which customizations actually land interviews.

Frequently Asked Questions

Does every ATS system work the same way?

No, and that gap matters. Enterprise platforms like Workday, Greenhouse, and iCIMS now layer semantic matching on top of exact-match scoring, so "led cross-functional teams" may get partial credit for "team leadership." But exact phrasing still carries more weight, and older or smaller systems may do strict string matching only. The safest approach is always to mirror the job description's exact language and treat semantic matching as a bonus, not a fallback.

How often should you update your resume for different jobs?

Every application, at minimum for the job title and Tier 1 keywords. A full rewrite isn't the point — swapping five to eight terms and mirroring the exact job title from the posting can move you meaningfully up the ranking without touching the rest of the document. Think of it as a ten-minute pass per application, not a two-hour project.

What file format is actually safest for ATS submission?

.docx is the safest default when the posting doesn't specify. Standard PDFs work fine on modern platforms, but design-tool exports — anything built in Canva, InDesign, or similar — use embedded text boxes and custom fonts that break parsing even as a PDF. If you're unsure whether your PDF is "standard," use .docx. Save design portfolios for a separate attachment only if the role specifically asks for one.

Are skills sections still worth including, or do ATS systems ignore them?

Skills sections still matter, but they matter less than where else your keywords appear. A keyword in a work experience bullet — showing you actually used the skill — ranks higher than the same word sitting alone in a list. The skills section is most useful for covering tools and certifications that don't fit naturally into a bullet, and for making sure the exact acronym and spelled-out form both appear on the document.

How long should an ATS-optimized resume be?

For early-career candidates, one page is the standard. For candidates with five or more years of experience, two pages is acceptable and often better — cutting real accomplishments to hit one page can strip out the specific detail and metrics that both the ATS and the human reviewer are looking for. What matters more than length is density: every line should carry a keyword, a number, or a named tool. Filler lines about "excellent communication skills" waste space that could hold evidence.