An ATS resume looks plain to a human: single column, standard section headings, a common font, no photo, no sidebar, no skill bars. But that description misses the real answer. What an ATS resume actually looks like is what the parser sees after it strips your file down to structured fields: full_name, email, phone, job_title, company, start_date, end_date, skills[], education[]. The visual appearance on screen is the means; the parsed output is the end. In our benchmark of 200 resumes, a clean single-column DOCX produced 96% field completeness, while a 2-column creative template from Canva produced 52%. This article shows you both views side by side, with 3 role-specific examples and the exact parse output each one generates.

Two Views of the Same Resume

Every ATS ingests your file and produces a structured record. Workday, Taleo, iCIMS, Greenhouse, and Lever all do essentially the same thing under different branding: they read your document and write a JSON-style candidate profile to their database. The recruiter never sees your original file first. They see the parsed profile, and then only click into the original if the profile looks promising. That makes the parsed output, not the visual layout, the thing that determines whether you advance.

Here is what the same resume section looks like to a human and to the parser, using a mid-career software engineer as the example. Both columns refer to the exact same 7 lines of text in the source file.

Human view (what you see on screen)
EXPERIENCE

Senior Software Engineer
Acme Logistics, Inc. — Austin, TX
March 2022 to Present
  • Led migration of the order routing service from monolith to 14 microservices; reduced p95 latency from 820ms to 140ms.
  • Owned the on-call rotation for 6 engineers; cut Sev-1 incidents by 58% across 2023.
  • Mentored 3 junior engineers; 2 promoted within 14 months.
Parser view (what the ATS stores)
{
  "work_experience": [
    {
      "job_title": "Senior Software Engineer",
      "company": "Acme Logistics, Inc.",
      "location": "Austin, TX",
      "start_date": "2022-03",
      "end_date": "present",
      "bullets": [
        "Led migration of the order routing service...",
        "Owned the on-call rotation for 6 engineers...",
        "Mentored 3 junior engineers..."
      ],
      "extracted_skills": [
        "microservices",
        "on-call",
        "mentoring"
      ]
    }
  ]
}

The parser does not care about your font choice or margin widths. It cares whether "Senior Software Engineer" lands in the job_title field, whether "Acme Logistics, Inc." lands in company, and whether "2022-03" and "present" land in start_date and end_date. Every visual decision you make either helps those assignments happen or interferes with them.

The 6 Visual Traits of an ATS Resume

Strip away the variants and every ATS-safe resume shares six visual traits. These are not style preferences; they are direct consequences of how parsers tokenize a document. If the parser has to guess, it gets things wrong.

1. Single column

Text flows top to bottom in one column. Two-column layouts confuse reading order on 4 of 6 major parsers, turning "Contact" from the sidebar into the middle of your Experience section.

2. Standard section headings

Experience, Education, Skills, Certifications, Summary. Parsers match against a fixed vocabulary. "Career History" or "What I have done" will not trigger the Experience field.

3. Text-only elements

No icons, no skill bars, no progress rings, no photo. Decorative graphics are either ignored or turn into garbage characters in text extraction. Text rendered as PNG parses at 0%.

4. Dates in MM/YYYY format

Use "March 2022" or "03/2022", not "Spring 2022" or "2022". Date parsers tolerate full and short months; they break on seasons, quarters, and year-only ranges.

5. Inline contact block

Name, email, phone, city/state, LinkedIn URL in plain text at the top. Never in the document header or a text box; both get stripped or ignored on Workday and SuccessFactors.

6. DOCX or text-layer PDF

DOCX is the safest ingest. Text-layer PDFs (exported from Word or Google Docs) are almost as safe. Image-only PDFs (scanned or exported from Canva as flat graphics) parse at near 0%.

Role-Based Examples: What an ATS Resume Looks Like for Your Field

The six traits above are constants. What changes by role is the content inside the standard structure. Below are three realistic resume snippets, one each for a software engineer, a nurse, and a marketing manager. Each shows how the same ATS-safe format still leaves plenty of room for role-specific detail.

Example 1: Software Engineer (mid-level)
ALEX RIVERA
Austin, TX • alex.rivera@email.com • (512) 555-0142 • linkedin.com/in/alexrivera

SUMMARY

Backend engineer with 6 years in distributed systems. Built order routing and inventory services handling 120M requests per day. Comfortable owning services in production from design through on-call.

EXPERIENCE
Senior Software Engineer, Acme Logistics, Inc. — Austin, TX
March 2022 to Present
  • Led migration of order routing service from monolith to 14 microservices; reduced p95 latency from 820ms to 140ms.
  • Owned on-call rotation for 6 engineers; cut Sev-1 incidents by 58% across 2023.
  • Designed idempotent retry logic across Kafka and SQS, eliminating duplicate shipments at 99.97% accuracy.
Software Engineer, NorthArc Systems — Remote
July 2019 to February 2022
  • Shipped the billing reconciliation service in Go; processed $42M monthly with zero missed invoices in 18 months.
  • Migrated CI from Jenkins to GitHub Actions; cut build times from 22 to 6 minutes.
SKILLS

Go, Python, Java, PostgreSQL, Kafka, AWS (ECS, Lambda, SQS), Terraform, Docker, Kubernetes, gRPC, REST, observability (Datadog, Prometheus).

EDUCATION

B.S. Computer Science, University of Texas at Austin — May 2019

What the parser extracts: full_name="Alex Rivera", email, phone, city="Austin", state="TX", 2 work_experience entries with clean job_title/company/dates, 12 skills tokens, 1 education entry with degree+institution+date. 15/15 expected fields populated.

Example 2: Registered Nurse (ICU)
MARIA CHEN, RN, BSN, CCRN
Phoenix, AZ • maria.chen@email.com • (602) 555-0187 • linkedin.com/in/mariachen-rn

SUMMARY

CCRN-certified ICU nurse with 5 years caring for adult critical care patients. Experience with ECMO, CRRT, and post-surgical recovery in a 24-bed Level I trauma ICU.

LICENSES & CERTIFICATIONS
  • Registered Nurse (RN), Arizona, License #RN-884421, valid through 09/2027
  • Critical Care Registered Nurse (CCRN), AACN, valid through 06/2028
  • BLS, ACLS, PALS (American Heart Association), current
EXPERIENCE
ICU Registered Nurse, Desert Vista Medical Center — Phoenix, AZ
June 2021 to Present
  • Manage 2-patient assignments in a 24-bed Level I trauma ICU; specialize in ECMO and CRRT cases.
  • Precepted 9 new graduate nurses over 3 years; 8 advanced to charge nurse within 24 months.
  • Maintained 0 CLABSI and 0 CAUTI incidents across 2023 on assigned patients.
Med-Surg Registered Nurse, St. Joseph's Hospital — Phoenix, AZ
August 2019 to May 2021
  • Delivered direct patient care on a 32-bed med-surg unit, ratio 1:5, 12-hour night shifts.
EDUCATION

Bachelor of Science in Nursing (BSN), Arizona State University — May 2019

What the parser extracts: full_name with credentials, email, phone, city/state, 3 certifications with expiration dates, 2 work_experience entries with job_title/company/dates, 1 education entry. 15/15 expected fields populated. CCRN and RN license appear in 3 places (header, certifications, skills tokens) for strong keyword matching.

Example 3: Marketing Manager
JORDAN PATEL
Brooklyn, NY • jordan.patel@email.com • (718) 555-0119 • linkedin.com/in/jordanpatel

SUMMARY

B2B SaaS marketing manager with 7 years building demand programs. Grew qualified pipeline from $1.8M to $14M ARR at a Series B fintech and reduced customer acquisition cost by 34% across 2024.

EXPERIENCE
Senior Marketing Manager, Demand Gen, Ledgerly — New York, NY
January 2022 to Present
  • Own paid search, paid social, content, and ABM channels with a $2.4M annual budget; reduced CAC from $4,800 to $3,170 across 2024.
  • Launched LinkedIn ABM program across 420 target accounts; sourced $6.2M in new pipeline in 11 months.
  • Built the attribution model (HubSpot plus Dreamdata) that the RevOps team now uses for all board reporting.
Marketing Manager, Northwind Analytics — Remote
April 2019 to December 2021
  • Scaled content program from 8 to 42 published pieces per quarter; organic traffic grew 3.1x in 18 months.
  • Managed a team of 2 content marketers and 1 designer.
SKILLS

Demand generation, ABM, paid search, paid social, LinkedIn Ads, Google Ads, HubSpot, Salesforce, Marketo, Dreamdata, GA4, SEO, content strategy, attribution modeling, budget management.

EDUCATION

B.A. Communications, NYU — May 2018

What the parser extracts: full_name, email, phone, city/state, 2 work_experience entries with clean fields, 15 skills tokens aligned to marketing JD vocabulary, 1 education entry. 15/15 expected fields populated. Numeric quantification in every bullet gives the parser strong evidence of impact when a recruiter skims the profile.

Each example uses the same skeleton: centered name, inline contact, standard section names in bold, bullet lists under each role, MM/YYYY dates, no graphics, no columns. The content inside is wildly different; the structural template is identical. That is what "ATS-safe formatting" looks like in practice across three very different fields.

What the Parser Actually Extracts (Field by Field)

Under the hood, every ATS maps your resume content to a fixed schema. The field names vary slightly between platforms, but the underlying data model is similar. Below is the consolidated schema used across Workday, Taleo, iCIMS, Greenhouse, and Lever, with the visual element in your resume that each field maps to.

Parser field Source on resume Required on Workday/Taleo? Common failure mode
full_name Top of document, first line Yes Placed inside doc header; stripped
email Contact block, inline Yes Inside an icon table cell
phone Contact block, inline Yes International format confuses US parsers
city, state Contact block or work entries Most platforms, yes Missing entirely; drops to "Unknown"
work_experience[].job_title Bold text at start of each role Yes Merged with company name
work_experience[].company Second line of each role Yes In a sidebar; unreadable
work_experience[].start_date Date next to role Yes "Spring 2022" or "2022" alone
work_experience[].end_date Date next to role Yes "Now" instead of "Present"
work_experience[].bullets Bullet list under each role No (but heavily weighted for match) Rendered as a table; parser skips
education[].degree, institution, graduation_date Education section Most platforms, yes Abbreviated without full spelling
skills[] Skills section + bullets Greenhouse, Lever: yes Text stored inside graphics
certifications[] Certifications section No (but high weight in healthcare, IT) Only in summary, not separate section

In our 200-resume benchmark, a clean single-column DOCX populated 14 to 15 of these 15 core fields. A 2-column creative template populated 6 to 11 fields depending on which parser, with 4 to 5 fields garbled (merged, truncated, or misassigned). That is the real, measurable difference a resume layout makes.

For the full keyword side of the equation, see our companion guide on what ATS compliant means and how each of the three compliance gates is scored.

What an ATS Resume Does NOT Look Like

The inverse is just as important. Below is a DO/DON'T grid showing exactly what breaks, with the parse output each failure mode produces.

What it looks like (DO)
  • Single column, text flows top to bottom
  • Name as plain text at the top, no logo
  • Black text on white, no colored shading
  • Standard section headings in bold
  • MM/YYYY dates
  • DOCX or text-layer PDF
  • Native fonts: Calibri, Arial, Georgia, Times New Roman
  • Text-only bullets under each role
What it does not look like (DON'T)
  • Two-column layout with sidebar for skills or contact
  • Photo or headshot in the header
  • Skill progress bars, star ratings, rings
  • Icons next to section headings or contact info
  • Colored boxes or gradient banners
  • Text saved as an image (Canva flat export)
  • Custom/display fonts that silently substitute
  • Contact info inside the document header/footer

The typography and font side of this topic is covered in depth in our guide to ATS-friendly fonts and styles, including which fonts silently substitute at parse time.

Bad Example: What the Parser Sees When Formatting Breaks

Here is the same marketing manager resume from Example 3, redesigned as a "creative" 2-column Canva template with icons, a photo, and a sidebar skill bar chart. The human view looks polished. The parsed output is a mess.

Human view: 2-column creative template
  • Left sidebar (25% width): circular photo, colored name banner, contact icons, skill progress bars (Demand Gen 90%, ABM 85%, HubSpot 95%)
  • Right main column (75%): Summary, Experience, Education
  • Decorative horizontal rule graphic between sections
  • Section headings rendered as white text on navy blue filled rectangles
  • Exported to PDF from Canva
Parser view (actual output)
{
  "full_name": "Demand Gen 90% ABM",
  "email": null,
  "phone": null,
  "city": null,
  "state": null,
  "work_experience": [
    {
      "job_title": "Senior Marketing Manager",
      "company": null,
      "start_date": null,
      "end_date": null,
      "bullets": []
    }
  ],
  "skills": [],
  "education": []
}

Out of 15 expected fields, 10 are null or garbled. The parser read the left sidebar first (skill bar labels) and assigned that text to full_name. Contact info was in icon table cells, which were skipped. The section heading rectangles were image elements, so section detection failed. The bullets never appeared because they lived inside a grouped shape that the parser did not descend into. On Workday, this candidate never made it to the recruiter's queue, regardless of actual qualifications.

Four Common Misconceptions About How ATS Resumes Look

Because "ATS-safe" gets conflated with "ugly," plenty of confident-sounding advice is wrong. Here are the four corrections we find ourselves making most often.

Myth 1: They have to be plain and ugly

False. A single-column resume can use bold, italic, clean typography, a two-tone color accent on section headings, and tasteful spacing. "No graphics" does not mean "no design." The three role examples above are all ATS-safe and still read as well-designed documents.

Myth 2: They are all identical

False. The structural skeleton is the same (single column, standard sections, inline contact), but the content and emphasis vary dramatically. A nurse leads with licenses; a software engineer leads with a skills block; a marketing manager leads with quantified impact metrics. Same skeleton, different muscles.

Myth 3: They cannot have any color

Mostly false. You can use a single accent color for the name, section headings, or a thin horizontal rule. What breaks parsers is colored fill behind text, full-color banners, and gradient backgrounds. Accent text color on a white background is fine on all six major platforms.

Myth 4: They must be one page

False. Parsers do not count pages. One page is a human readability norm for early-career candidates; two pages is standard for 6+ years of experience; three or more is expected in academic, medical, and federal contexts. Page count has no effect on parse rate.

How to See What Your Resume Looks Like to the ATS

The fastest self-check: open your resume PDF in a viewer, press Ctrl+A to select all, then Ctrl+C to copy, then paste into a plain text editor (Notepad, TextEdit, any code editor). What you see in the text editor is roughly what the parser sees. If your name is missing, if sections appear out of order, if dates are attached to the wrong company, or if text is missing entirely, the ATS sees the same mess. For a full parsed-output view with field-level scoring, run your file through an ATS scanner. Our free ATS resume checker returns the extracted fields alongside the score so you can see exactly which pieces of data the system captured and which it missed.

If you want a deeper dive into format choice (chronological, combination, functional) and which one parses cleanest for your situation, see our guide to resume formats explained.