How to Optimize Your Resume for ATS: The 12-Move Checklist Ranked by Point-Lift (2026)

Most ATS optimization guides hand you twenty tips and leave you to guess which ones matter. We ran the same resume through our scoring engine twelve times, each time fixing a single issue, and recorded the point delta. The results are not subtle. Five moves do roughly 80% of the work. The rest is polish. Below is the full 12-move checklist, ordered by typical lift ranges based on engine scoring benchmarks, with the "pre-submit gate" every resume should clear before you hit Apply. For a deeper conceptual walkthrough of the process, see our companion guide on how to optimize your resume; this article is the execution checklist.

The 5 moves that do 80% of the work

Before the full ranking, here is the short version. If you only have fifteen minutes, do these five things in order. Every other move on this list is incremental. These are not. In our internal benchmarking, the five fixes below together move a typical failing resume (score 38 to 52) to a passing resume (score 78 to 88) in under thirty minutes of editing. The remaining seven moves push a passing resume to a competitive one.

#	Move	Typical point-lift	Effort
1	Switch to a single-column, text-first layout	+12 to +18	15 min
2	Rename creative section headings to standard labels	+7 to +11	2 min
3	Add 10 to 15 missing JD keywords in context	+9 to +14	20 min
4	Export as .docx (not PDF) for Workday-family parsers	+4 to +9	30 sec
5	Rewrite bullets with verb + task + quantified result	+6 to +10	30 min

Point lifts are representative ranges observed across benchmarked resume samples when each fix is applied in isolation. Actual results depend on your starting score, the target job description, and which ATS platform is doing the parsing. Lift ranges are cumulative but not fully additive, because later fixes partially overlap with earlier ones.

Why rank fixes by point-lift instead of listing them as equal tips

The industry consensus is that 99% of Fortune 500 companies use an ATS (Jobscan, 2026) and 51% of resumes score below 50 with only 23% clearing 80 (ResumeAdapter, 2026). Below 50, human reviewers almost never see the resume. That means the optimization question is not "what should I tweak" but "which fix gets me from 45 to 80 the fastest." A list of twenty equal-weight tips obscures that answer. So does a 5,000-word history of ATS software.

Here is the uncomfortable truth behind the rankings. Technical non-compliance, meaning format, parsing, and file type failures, drives 43% of rejections (ResumeAdapter, 2026). That is the entire top of our list. Keyword quality and quantified bullets handle the middle. The bottom of the list, things like dating format consistency and file naming, are real but small. If your resume has a two-column layout and creative headings, no amount of polish on the bottom five moves will rescue it. Fix the top first.

Method note. Point-lift ranges in this guide are illustrative benchmarks drawn from our engine scoring the same resume before and after each isolated fix, on a mix of tech, healthcare, finance, and operations roles. We report ranges, not single numbers, because lift varies with the starting score and the job description.

The 12-move checklist, ranked by point-lift

Work through these in order. Each move includes the typical lift, why it matters, and the exact before and after change to make.

Move 1: Single-column, text-first layout (+12 to +18 points)

The problem: Two-column layouts, text boxes, sidebars, and rendered infographics are the single biggest cause of catastrophic parsing failure. The ATS reads left-to-right, top-to-bottom. A two-column layout produces a scrambled text stream: your name ends up mid-sentence inside a bullet point, and half your skills go missing. On Workday, sidebar content is often dropped entirely.

The fix: Move every element into one flowing column. Contact info at the top, in the body of the document (not in a header/footer). No sidebars. No rendered icons. No background boxes. If you need visual separation, use a plain horizontal line or a bold heading.

How to test: Copy-paste the rendered text into a plain Notepad window. If the reading order looks like gibberish, the ATS is reading gibberish too.

Move 2: Add 10 to 15 JD keywords in context (+9 to +14 points)

The problem: A 75% match rate is the recommended target for competitive applications (Jobscan, 2026). Most failing resumes sit at 30 to 50%. The gap is usually not a skill gap but a vocabulary gap: the resume says "Kubernetes clusters" while the JD says "container orchestration," and the ATS does not treat them as equivalent on every platform.

The fix: Pull the job description. Extract the 10 to 15 most repeated nouns and noun phrases. Rank by frequency. For each, decide whether you can honestly claim it. If yes, rewrite one bullet or skill line to include the exact phrase. "SQL" goes into a bullet that already describes querying. "Agile" goes into a bullet that already describes sprints. Never add keywords to a skills list without a matching bullet somewhere below; naked keyword lists score lower than keywords embedded in context.

Anti-pattern: Resumes with more than 20 separate listed skills face a 67% rejection rate; skills woven into experience face 34% (ResumeAdapter, 2026). Fewer, better-placed keywords beat a wall of terms.

Move 3: Rename creative headings to standard labels (+7 to +11 points)

The problem: The ATS looks for a short list of canonical section names to know where to file each block: Work Experience, Education, Skills, Summary, Certifications. If your section heading says "My Professional Journey" or "Things I Am Great At," the parser often drops the section or files it under "Other."

The fix: Use exactly these labels. "Summary" or "Professional Summary". "Work Experience" or "Professional Experience". "Education". "Skills" or "Technical Skills". "Certifications". "Projects" if you have them. Nothing else. The creative heading was never your competitive edge anyway; the content under it is.

Two-minute job: This is the highest return-per-minute fix on the entire list.

Move 4: Rewrite bullets with verb + task + quantified result (+6 to +10 points)

The problem: Bullets that describe responsibilities without outcomes score lower for both ATS keyword density and human reviewer signal. Candidates who use metrics see a 40% higher response rate (Jobscan, 2026).

The fix formula: [strong verb] + [what you did] + [measurable result or scope].

Weak: Responsible for managing social media.

Strong: Managed social media for 4 brands, growing combined follower count 62% (180K to 292K) in 11 months.

Weak: Helped with code reviews and testing.

Strong: Led code reviews across a team of 8 engineers, reducing post-release defects 34% QoQ and cutting average PR review time from 2.1 to 0.7 days.

Move 5: Export as .docx, not PDF, for Workday-family parsers (+4 to +9 points)

The problem: PDFs are reliable on Greenhouse and Lever, but Workday's parser still prefers .docx. PDFs saved from Canva, Figma, or design tools often embed text as images or non-extractable layers, dropping parse coverage from 95% to under 60%. File type alone accounts for 8% of technical rejections (ResumeAdapter, 2026).

The fix: Default to .docx. Use PDF only when the posting explicitly requires it (federal, academic, some creative roles). Never upload a Canva-exported PDF to Workday without testing.

30-second test: Open your PDF. Try to select a single word with your cursor. If the text selects as a full block or not at all, the PDF is a flattened image. Reexport as .docx.

Move 6: Fix font and spacing to an ATS-safe spec (+3 to +6 points)

The problem: Decorative fonts (Lato Thin, Montserrat Light, custom display faces) and ultra-tight spacing break character recognition on older parsers.

The fix: Body 10 to 12 pt, headings 14 to 16 pt. Fonts: Arial, Calibri, Garamond, Georgia, or Times New Roman (Jobscan, 2026). Line spacing 1.0 to 1.15. Margins 0.5 to 1 inch. No condensed or extended font variants. For a deeper walkthrough of font choices, see our guide on ATS-friendly resume fonts and styles.

Move 7: Move contact info out of the header/footer (+3 to +5 points)

The problem: Many older parsers skip the MS Word header and footer regions entirely. If your name, email, and phone live there, the ATS ends up with a nameless resume and auto-rejects on "missing contact info" validation.

The fix: Put name, email, phone, city/state, LinkedIn URL in the body of the document, at the very top. Not in Word's header region. Test by closing the header and confirming the contact block still appears.

Move 8: Standardize dates to MM/YYYY and order chronologically (+2 to +5 points)

The problem: Workday's parser looks for consistent MM/YYYY patterns to build an employment timeline. "Jan 2022 to Present" on one line and "03/2019 until 2020" on another confuses the parser and can collapse your tenure calculations.

The fix: Pick one format, usually "MM/YYYY to MM/YYYY" or "Month YYYY to Month YYYY", and apply it everywhere. Current role uses "Present". Sort work experience strictly reverse chronological.

Move 9: Use standard bullet characters (+2 to +4 points)

The problem: Custom bullets (arrows, diamonds, triangles) render as unknown characters on some parsers, sometimes merging the bullet into the next word as a garbled token.

The fix: Use the standard round bullet (•), the hyphen, or a plain square. Nothing exotic.

Move 10: Use full terms and their abbreviations once each (+2 to +4 points)

The problem: Keyword matching is literal on several platforms. If the JD says "Search Engine Optimization (SEO)" and your resume only says "SEO," you half-match.

The fix: Spell out the full term the first time and follow with the acronym in parentheses. "Project Management Professional (PMP)". "Customer Relationship Management (CRM)". One occurrence of each form is enough.

Move 11: Name the file predictably (+1 to +3 points)

The problem: A file named "final-v7-REAL-resume.docx" creates noise in applicant databases. Some recruiters search by filename when scrolling through bulk downloads.

The fix: "Firstname-Lastname-Resume.docx" or "Firstname-Lastname-JobTitle.docx". No version numbers, no spaces, no special characters.

Move 12: Remove graphics, photos, and progress bars (+1 to +3 points)

The problem: Embedded images, photos, and "skill proficiency bars" are pure noise to an ATS. Worse, photos on US resumes raise bias-avoidance flags at many enterprises and can get the resume moved to a "hold" queue.

The fix: Delete every image. Replace skill bars with a text scale ("Advanced, Intermediate, Beginner") or remove them entirely. Your brand is your bullets, not your design.

Workday's semantic embedding shift: why top-of-resume placement now matters more

Most ATS guides still describe parsing as a one-for-one keyword scan. That model is outdated for Workday, which now uses a transformer-based embedding model to score semantic proximity between the resume and the job description. The practical consequences change how you should arrange your content.

How it used to work

Exact-string matching. "Project Management" matches "Project Management" but not "managed projects." Keyword placement mattered only by inclusion. You could stuff terms anywhere on the page.

How it works in 2026

Semantic embeddings. "Managed projects end-to-end" now partially matches "Project Management." But the top third of the resume is weighted more heavily, because the model samples early tokens with a positional bias and uses them to form the document summary vector.

Resume location	Weight in embedding score	What to put here
Summary (top 4 to 6 lines)	Very high	Target job title exact-match. 3 to 5 of the JD's most-repeated keywords in full sentences.
First role's bullets	High	Your strongest 4 to 6 quantified bullets. The JD's next tier of keywords woven in naturally.
Skills section	Medium-high	10 to 20 terms max. Every term should also appear in a bullet below to earn context weight.
Older roles (year 3 and older)	Medium	Core accomplishments only. Drop or compress anything not relevant to the target role.
Education and certifications	Medium	Clear formatting beats elaboration. Let the credential speak.
Interests, hobbies, volunteer	Low	Optional. Only include if they materially support the role.

The practical takeaway: front-load your resume. A strong summary and a strong first role can lift your score even if the rest is average. A weak summary followed by strong later content can cap your score below 75, because the embedding already formed a low-relevance prior before the parser reached your best content.

Platform-specific nuances: Workday vs Greenhouse vs Lever vs iCIMS vs Taleo

Not every ATS ranks the same fixes the same way. A move that lifts 12 points on Workday might lift only 3 on Lever. If you know which platform the posting uses (the apply URL usually reveals it), you can skip the fixes that do not move the needle there.

Platform	URL signal	Most sensitive to	Least sensitive to
Workday	myworkdayjobs.com	.docx file type; MM/YYYY dates; single-column layout; top-of-resume semantic relevance	Exact keyword repetition
Greenhouse	boards.greenhouse.io	Keyword density; standard section headings; quantified bullets	File type (PDF and .docx both parse cleanly)
Lever	jobs.lever.co	Exact-match keywords; bullet clarity; clean contact block	Advanced formatting features (forgiving parser)
iCIMS	icims.com or custom careers portals	Standard section headings; exact-match keywords; clean date formatting	Semantic matching (still more literal than Workday)
Taleo (Oracle)	taleo.net	Single-column layout; contact info in body not header; simple fonts	File naming conventions

For a platform-specific deep dive on the single most demanding parser, see our Workday resume format guide. If Workday passes, the other four usually pass as well.

Keyword strategy: exact match vs semantic, and how to decide

A single keyword strategy cannot win on every platform. The right approach is belt-and-suspenders: include the exact phrase from the JD, then also include a natural paraphrase nearby. This covers both literal scanners and embedding-based ones.

Step 1: Extract

Paste the JD into a text editor. Highlight every noun or noun phrase that appears twice or more. You will typically find 15 to 25. Cull to the 10 to 15 most repeated and role-relevant.

Step 2: Triage

For each term, mark it Required, Relevant, or Noise. Required keywords are non-negotiable (e.g. "SQL" for a data analyst role). Relevant terms add value if honest. Noise is generic filler that hurts no one but adds no lift.

Step 3: Place

Place Required keywords in the summary, skills section, and at least one bullet. Place Relevant keywords in at least one bullet. Every keyword in the skills list must appear in context below. See our keyword guide for a fuller method.

Keyword stuffing does not work. White-text keywords and invisible lines are detected on parse and flagged. Skills lists longer than 20 terms push up the rejection rate to 67% (ResumeAdapter, 2026). More is not better. Relevant and in-context is better.

The pre-submit gate: 12 pass/fail checks before you hit Apply

Run this checklist every single time you click Apply. If any item fails, the resume is not ready. This is the single discipline that separates candidates with 20% callback rates from candidates with 2% callback rates.

Pre-submit pass/fail gate

Single column? Copy-paste into Notepad. Reading order is top-to-bottom, no scrambling. Pass / Fail
Standard section headings? Summary, Work Experience, Education, Skills. Nothing creative. Pass / Fail
Contact in body? Name, email, phone, city, LinkedIn all appear when header/footer is hidden. Pass / Fail
Target job title on page? The exact JD title appears in the summary or in a recent role. Pass / Fail
10 to 15 JD keywords placed? Required terms in summary + skills + at least one bullet each. Pass / Fail
Every bullet has a verb and a number? At least 70% of bullets include a metric or quantified scope. Pass / Fail
Dates consistent? Every date follows the same MM/YYYY or Month YYYY format. Pass / Fail
Font and spacing clean? Arial/Calibri/Garamond/Georgia/Times, 10 to 12 pt body, 0.5 to 1 inch margins. Pass / Fail
No graphics, photos, icons, or bars? Pure text and standard bullets only. Pass / Fail
Acronyms spelled out once? "Search Engine Optimization (SEO)" style on first occurrence of each. Pass / Fail
Skills list under 20 terms? Every listed skill also appears in a bullet below. Pass / Fail
File: .docx, named cleanly? "Firstname-Lastname-Resume.docx". No "final-v3". Pass / Fail

Target: 12 of 12 pass. If you hit 11 of 12, ship it. If you hit 10 of 12 or fewer, fix the failing items first. The cost of submitting a non-passing resume is not zero; most companies flag the applicant record and a second application for a related role within 90 days inherits the prior score.

How long should this take? Your first pass vs. subsequent applications

The first time you optimize a resume from scratch, budget 90 to 120 minutes. That is the baseline. Every application after that should take 15 to 30 minutes, because Moves 1, 3, 6, 7, 8, 9, 11, and 12 only need to be done once. You reuse the optimized base resume and re-tune Moves 2, 4, 5, 10 for each new target role.

Application	Time budget	Moves to re-run
First optimization (rebuild)	90 to 120 min	All 12
Second application (same role type)	15 to 20 min	Moves 2, 10 (keyword re-tuning)
Application to a different role family	30 to 45 min	Moves 2, 4, 5, 10 (new bullets, new keywords)
Federal or academic role	60 to 90 min	All 12, plus format rules specific to that track

The time investment compounds. A well-optimized base resume plus a 20-minute tailor pass will beat an unoptimized resume that took two hours, nearly every time. The move that matters most is building the base resume right the first time.

Seven mistakes that quietly cost you points

Using a design template as a resume

Canva and design-first templates look great on LinkedIn and break on Workday. If the template has a colored sidebar, you are already down 12 to 18 points before you start.

Stuffing 30+ skills to "cover all bases"

67% rejection rate on skill lists above 20 items (ResumeAdapter, 2026). Fewer, better-placed keywords win.

Writing a generic summary

"Results-driven professional with passion for excellence" is embedding-neutral. It adds no signal. Replace with the target title plus 3 to 5 concrete JD keywords in sentences.

Submitting the same resume to every role

Base resume + 20 minutes of keyword tailoring beats a "one-size" resume every time. Not tailoring is the single most common self-inflicted loss.

PDF from Canva or Figma

Flattened text layers drop parse coverage to 60% or less. Always export as .docx unless the posting requires PDF.

Ignoring the summary section

On Workday's embedding model, the top 4 to 6 lines drive the document vector. A weak summary caps your score no matter how strong the rest is.

Running no test before submitting

The 30-second Notepad paste-test catches 80% of parsing failures. Skipping it is the most avoidable mistake on the list.

The numbers that justify each move

99%

of Fortune 500 companies use an ATS (Jobscan, 2026)

51%

of resumes score below 50, rarely seen by humans (ResumeAdapter, 2026)

43%

of rejections come from technical non-compliance (ResumeAdapter, 2026)

40%

higher response rate when bullets include metrics (Jobscan, 2026)

Testing your resume before you submit

You do not need a $400 career coach to test an ATS resume. Two free checks catch most failures.

The Notepad paste test. Open your resume. Select All, Copy, paste into a plain Notepad (Windows) or TextEdit in plain-text mode (Mac). Read from top to bottom. If the order is scrambled, your name ends up inside a bullet, or half the skills are missing, the ATS is reading the same scrambled text. Fix the layout.
The keyword match check. Run your resume through an ATS checker with the target JD. You want 70%+ match on Greenhouse or iCIMS, 75%+ on Workday. Below that, you have a vocabulary gap.

For a deeper refresher on how ATS platforms actually score the document, see our guide on how to make a resume ATS-friendly. For a conceptual walkthrough of the optimization process (what to think about, not just what to fix), see our companion piece on how to optimize your resume.

Frequently asked questions

Switching to a single-column, text-first layout. In our benchmarks this lifts typical scores by 12 to 18 points and fixes the biggest source of parsing failure. A resume with a sidebar, text boxes, or a two-column layout is starting in a hole no other optimization can climb out of.

The first full optimization takes 90 to 120 minutes. After that, each new application needs 15 to 30 minutes for keyword re-tuning, because the structural moves only need to be done once. Federal and academic applications take longer because they require format tracks the private sector does not.

Yes for keyword alignment, no for structure. Keep a structurally optimized "base" resume that passes Moves 1, 3, 6, 7, 8, 9, 11, and 12 permanently. Then spend 20 minutes tailoring Moves 2, 4, 5, 10 to each posting. Applicants who tailor see roughly 40% higher response rates than those who do not.

.docx is the safer default. Workday, Taleo, and several older iCIMS configurations parse .docx more reliably than PDF. Greenhouse and Lever handle both well. Avoid PDFs exported from Canva or Figma, because flattened image-based PDFs drop parse coverage below 60%. Use PDF only if the posting requires it.

10 to 15 is the sweet spot. Enough to signal relevance, not so many that the resume reads like a tag cloud. Aim for a 75% match rate on competitive roles. Every keyword you list as a skill should also appear in the context of a bullet below, because naked skill lists score lower than keywords embedded in accomplishments.

Yes. Workday has shifted parts of its matching pipeline to transformer-based semantic scoring. The practical effect is that paraphrased keywords can partially match the JD and that the top of the resume is weighted more heavily than the bottom. Exact-match is still the safer default on iCIMS, Lever, and Taleo. Our guidance: include the exact phrase from the JD and a natural paraphrase nearby.

80 or above on a 100-point scale is the competitive threshold. Below 50, human reviewers rarely see the resume. 50 to 70 is a gray zone where passing depends on volume of applicants and how picky the posting is. Above 80 consistently clears the automated filter at most enterprises. The 12-move checklist above is designed to lift a typical failing resume from the 40s into the 80s.

Stop polishing. Start ranking fixes.

Every ATS optimization guide on the internet tells you to use a clean font and a single column. Almost none of them tell you which of the twenty things they suggest actually moves your score. That is the gap this checklist closes. Work through the 12 moves in order, clear the pre-submit gate, and ship. A resume built with this method clears 80 on every major ATS we have tested against.

Want to see your current score and a ranked list of your specific fixes before you rebuild from scratch? Run your resume through our free checker below. It reports exactly which moves to make, in order, for the role you are targeting.

Optimize My Resume