How to Build an ATS-Friendly Resume From Scratch in 2026

Here's a fact that most job seekers don't fully reckon with: the majority of online job applications are never read by a human being.

Not because hiring managers are lazy. Because the volume is impossible to manage manually. A mid-sized company posting a software engineering role in 2026 can receive 400–800 applications within 72 hours of posting. A Fortune 500 company hiring for a popular role can receive several thousand.

No recruiting team manually reviews all of those. They use Applicant Tracking Systems — ATS software — to filter, score, and rank applications before a human ever opens a single file.

If your resume doesn't pass the ATS filter, the hiring manager never knows you applied. Your qualifications don't matter. Your experience doesn't matter. You're effectively invisible.

This guide explains exactly how ATS software works, the specific reasons resumes fail it, and how to build a resume that passes the filter and impresses the human on the other side.

A resume that looks beautiful in Figma but fails an ATS scan is a resume that gets you zero interviews. Format for the machine first. Design for the human second.

How ATS Software Actually Works

Understanding the technology changes how you approach your resume. ATS software is not magic — it's a parser and a scorer. Once you understand what it's doing, the "rules" stop feeling arbitrary.

Step 1: Parsing

When you submit a resume, the ATS parses it — converts the document into structured data it can read and store. It extracts:

  • Contact information
  • Work history (employer, title, dates, descriptions)
  • Education (institution, degree, dates)
  • Skills
  • Certifications
  • Any other sections it recognises

This is where most resumes fail first. If the parser can't correctly extract your information — because of formatting that confuses it — your data enters the system incorrectly or incompletely. A resume in a two-column layout may have the text scrambled. A resume with custom section headers the ATS doesn't recognise may misfile your experience. A PDF with embedded graphics may produce garbled text when parsed.

The ATS isn't wrong to fail your resume in these cases. It just can't read it.

Step 2: Keyword Scoring

After parsing, the ATS compares your resume text against the job description. It's looking for keyword matches — specific terms, phrases, job titles, skills, and qualifications that appear in the posting.

Most ATS systems score resumes on a 0–100 scale based on keyword density and relevance. Resumes below a threshold score (typically 60–70%) are automatically filtered out before any human review.

This is where the second major failure happens. If your resume uses different language than the job description — even for the same skills — the ATS may not recognise the match. "Front-end development" and "frontend engineering" may score differently. "Project management" and "programme management" may not cross-match. "Machine learning" and "ML" may not be treated as equivalent depending on the ATS.

Step 3: Human Review

Only resumes that clear the automated filter reach human review. At this stage, the recruiter is spending 7–10 seconds on the initial scan before deciding whether to read further. This is where design, readability, and presentation matter.

Your goal: optimise for the machine in steps 1 and 2, and for the human in step 3.

The 8 Reasons Resumes Fail ATS

1. Multi-Column Layouts

Two-column or three-column layouts are popular in resume templates because they look clean and compact. ATS parsers read documents left-to-right, top-to-bottom in a single stream. A two-column layout may cause the parser to merge text from both columns into a single stream, producing nonsense output.

Fix: Use a single-column layout for all content. Clean, clear, single-column.

2. Tables for Layout

Using a table to organise your skills into a neat grid looks good visually. ATS parsers often can't read table content correctly and may skip it entirely or mangle the text.

Fix: List your skills as comma-separated text in a plain paragraph or a simple bulleted list. No tables.

3. Headers and Footers

Contact information placed in a document header or footer is frequently skipped by ATS parsers. The parser reads the body text but may not process header/footer regions — meaning your name, email, and phone number may not be captured.

Fix: Put all contact information in the body of the document, at the top.

4. Graphics, Icons, and Images

A skills section with star ratings. Icons next to section headers. A profile photo. Infographic-style elements. All of these are images from the ATS perspective — and ATS parsers read text, not images. The content inside graphics is invisible to the system.

Fix: Remove all graphics, icons, photos, and image-based design elements. Text only.

5. Unusual Section Headers

Standard ATS systems recognise standard section headers: "Work Experience," "Education," "Skills," "Certifications," "Summary." Creative headers like "Where I've Made an Impact," "My Toolkit," or "The Story So Far" may not be recognised, causing the content beneath them to be misfiled.

Fix: Use conventional, recognisable section headers. Creativity in headers costs you in ATS scores.

6. Incorrect or Unusual File Formats

Most ATS systems read .docx and standard .pdf files. Not all PDFs are equal — PDFs exported from design tools like Figma, Adobe Illustrator, or Canva are often image-based PDFs where the text isn't actually selectable or parseable. They look identical to a word-processed PDF on screen but contain no machine-readable text.

Fix: Use .docx when in doubt — it's universally parseable. If using PDF, export from Word or Google Docs, not a design tool. Test your PDF by selecting and copying the text — if you can't select it, the ATS can't read it.

7. Keyword Gaps

Even a correctly formatted resume can score poorly if it doesn't include the terms the ATS is matching against. Generic job descriptions produce generic keyword matching. But targeted job applications require targeted keyword alignment.

Fix: See the keyword optimisation section below.

8. Excessive Formatting

Bold, italic, underline, all-caps — used sparingly, these are fine. Used excessively (bolding every bullet, italicising every company name, using custom colours), they can confuse parsers and look chaotic to the human reviewer on the other side.

Fix: Bold only your most important structural elements (section headers, job titles, company names). Use italics sparingly. No colour in the body of the document (a subtle colour for your name is typically fine).

Building the ATS-Friendly Resume: Structure

Contact Section (Top of Document, Body Text)

Include:

  • Full name (largest text on the page — this is one appropriate style choice)
  • Professional email address (firstname.lastname@gmail.com or a custom domain)
  • Phone number
  • City and province/state (full address not necessary)
  • LinkedIn URL
  • Portfolio URL — your linkspaghetti.com/yourname link here
  • GitHub URL (for technical roles)

All of this in plain text in the document body. No icons. No header region.

Professional Summary (Optional but Often Effective)

2–4 sentences that position you for the specific type of role you're targeting. Include your years of experience, your primary area of expertise, and your differentiating value.

For ATS purposes: Include 2–3 of the most important keywords from the target job description naturally in your summary.

Skip the summary if: You're entry-level with fewer than 2 years of experience, or if it would just repeat what's already in your experience section.

Work Experience (The Core)

Reverse chronological order (most recent first). For each role:

  • Job title (match the market-standard title, not necessarily your exact internal title — "Product Designer" rather than "Experience Architect")
  • Company name and location
  • Employment dates (Month Year – Month Year format, or "Present")
  • 3–6 bullet points describing your contributions and results

The bullet point formula: Action verb + what you did + measurable result

  • "Redesigned the onboarding flow, reducing drop-off rate by 34% and increasing Day-7 retention by 18%"
  • "Built a data pipeline processing 2M records daily, reducing latency by 60% versus the previous batch system"
  • "Led a cross-functional team of 8 to ship a mobile feature used by 400K active users within the first month"

Quantify wherever possible. If the exact number is sensitive, use a range ("reduced costs by 25–30%") or a relative comparison ("cut processing time in half").

Skills Section

A dedicated skills section serves two purposes: it gives the ATS a concentrated block of keywords to score, and it gives the human reviewer a quick-scan reference.

Format as a clean list of relevant skills, tools, and technologies. For technical roles, group by category:

  • Languages: Python, TypeScript, Go
  • Frameworks: React, Next.js, FastAPI
  • Tools: Figma, Jira, dbt, Snowflake

Keep this section accurate. Including skills you can't speak to fluently in an interview is a liability, not an asset.

Education

Degree, institution, graduation year. For recent graduates, include relevant coursework, academic projects, and GPA if above 3.5. For experienced professionals, keep education brief — one or two lines per degree.

Keyword Optimisation: The Right Way

Keyword stuffing — pasting lists of keywords into a white font, or cramming 40 skills into a tiny paragraph — is detectable by modern ATS systems and a red flag for human reviewers. The goal is intelligent alignment, not gaming.

Step 1: Analyse the job description

Read it carefully and identify:

  • The exact job title used
  • Technical skills mentioned (tools, languages, platforms)
  • Soft skills mentioned repeatedly ("cross-functional collaboration," "stakeholder management")
  • Industry-specific terminology
  • Required qualifications and how they're phrased

Step 2: Cross-reference against your resume

Highlight every term from the job description that also appears in your current resume. Note every important term that's missing from your resume but present in the posting.

Step 3: Add missing keywords naturally

Don't add keywords you don't have. Add keywords for skills you genuinely possess but described differently. If the job description says "Agile methodologies" and your resume says "sprint-based development," update your language to match the posting.

Step 4: Mirror the job title

If you're applying for a "Senior UX Designer" role and your resume says "Principal Experience Designer," consider whether the title difference is hurting your keyword match. Many ATS systems match job titles as part of their scoring. Your resume title in the summary section can reflect the role you're targeting rather than your exact internal title.

How MyJobTracker helps here: The resume builder in MyJobTracker lets you save multiple resume versions and tag each application with the version used. When you customise a resume for a specific role and want to track which tailored version produces better results, you need a system that connects the version to the outcome. This is exactly what the version-tracking feature does.

Testing Your Resume Before You Submit

Before submitting to a single application, test your resume against an ATS simulator:

Free tools:

  • Jobscan.co — Paste your resume and a job description and get a keyword match score plus specific gaps
  • Resume Worded — ATS scan plus line-by-line suggestions
  • LinkedIn's resume builder — Shows keyword alignment with saved jobs

Manual test: Open your resume PDF, select all text, and paste it into a plain text editor. If the output is readable and logically structured, the ATS can likely parse it correctly. If it's scrambled — columns merged, sections out of order, text missing — your formatting is problematic.

Target a keyword match score of 65–80% for any given job description. Below 65%, you're likely getting filtered out. Above 80%, you may be over-tailoring at the expense of readability.

The Human Side: What Happens After the ATS

Your resume has cleared the filter. A recruiter has 7–10 seconds before deciding to read further. What are they actually doing in those 7 seconds?

They scan the visual structure first. Is this clean and readable? Are the sections clearly defined and easy to navigate? Is the text density appropriate (not walls of text, not sparse to the point of being uninformative)?

They find your most recent role immediately. What's the company? What was the title? Is this someone who's been progressing? Are they from a credible organisation?

They scan for impact language. Bullet points that start with strong action verbs and contain numbers read as credible and competent. Bullet points that describe responsibilities without outcomes read as generic.

They look for anything that breaks the pattern. A typo, an unusual formatting choice, a conspicuous gap, a mismatch between the role title and the company's known reputation in the industry.

Your ATS-formatted resume, when structured with the single-column, keyword-aligned approach described above, is also exactly what human reviewers prefer to scan. The ATS-friendly format is also the human-friendly format.

The Resume Versions You Should Have

A single resume for every application is the job seeker's equivalent of sending the same pitch to every investor. Different companies, different role levels, and different industries warrant different framings of the same underlying experience.

Base version: Your fully optimised, general-purpose resume. Clean formatting, strong bullets, accurate skills section, portfolio link included.

Technical depth version: For roles at engineering-forward companies where the ATS and human reviewer both reward technical specificity. Lead with systems architecture, performance metrics, technical decisions made.

Impact-forward version: For business-focused companies or product roles where business outcomes matter more than technical methods. Lead with revenue influenced, cost reduced, users served, time saved.

Entry-level or career-transition version: If you're early career or pivoting, a version that leads with transferable skills and reframes experience through the lens of the new direction. More summary emphasis, more relevant projects featured.

Each version lives in MyJobTracker as a saved resume. Each application is tagged with the version used. After 20–30 applications per version, you have real data on which version is producing better results in your target market.

Your ATS Resume Checklist

Before any submission, confirm:

  • [ ] Single-column layout throughout
  • [ ] No tables, graphics, icons, or images
  • [ ] Contact information in the document body (not header/footer)
  • [ ] Standard section headers (Work Experience, Education, Skills, etc.)
  • [ ] File format: .docx or word-processed PDF (not design-tool PDF)
  • [ ] Job title in resume matches the target role title or common market title
  • [ ] Key terms from the job description are present naturally in the resume
  • [ ] All bullet points follow an action verb + contribution + result structure
  • [ ] Skills section is accurate and updated for this role
  • [ ] Portfolio link is present (every application, every time)
  • [ ] No typos (run spell check AND read aloud)
  • [ ] Paste-to-plaintext test passed (text is readable in plain format)
  • [ ] ATS simulator score is 65%+ for the target job description

The Bottom Line

ATS rejection isn't personal. It's mechanical. A resume that fails an ATS scan isn't a bad resume — it's a correctly formatted resume for human reading that was never adapted for machine reading.

The adaptation is straightforward, once you understand the rules. Single-column. No graphics. Standard headers. Keywords aligned to the job description. Portfolio link present. Test before you submit.

Pass the machine. Impress the human. Get the interview.

Get Started Today

MyJobTracker — Build ATS-ready resumes with the built-in resume builder, save multiple versions, and track which version generates better response rates across your applications. Free to start.

Build the resume that actually gets read.

Continue Reading: