ATS resume guide — what applicant tracking systems actually parse
Applicant tracking systems (ATS) are software tools that employers use to receive, store, and search job applications. They do not make hiring decisions. Understanding what they actually do helps you decide which formatting choices matter and which advice to ignore.
What an ATS parses
An ATS extracts text from your resume and attempts to sort it into fields: name, contact details, employment history, education, skills. The quality of the extraction depends on how you formatted the document.
Selectable text in a PDF or DOCX parses reliably. Text inside an image does not. Text in tables can parse but often ends up in the wrong field. Text in multi-column layouts is sometimes read in the wrong order. Unusual section headings can confuse the parser about which section is which.
What breaks parsing
The following formatting choices commonly break parsing:
- Text inside images, including your name if it is part of a header graphic.
- Text in tables, especially when the table defines the layout.
- Multi-column layouts where the reading order is ambiguous.
- Section headings that do not match standard names (use "Work Experience", not "Career Highlights").
- Date ranges without clear separators, such as "2019 2022" instead of "2019 – 2022".
- Non-standard file formats. PDF and DOCX are universally supported; other formats are not.
What does not break parsing
- Colour accents. A blue heading is still text.
- Serif fonts. Any widely available font parses fine.
- Single-column layouts with standard headings.
- Page margins from 12 to 25 millimetres.
- Unicode characters in your name and location, provided the font supports them.
Optimising without stuffing
Match the vocabulary of the job description. If the role says "stakeholder management", use that phrase. If it says "customer success", use that phrase. Do not try to force five synonyms into a single resume.
Be specific about tools and technologies. "Used Figma to design a mobile onboarding flow" is better than "Used design tools" and better than "Figma Figma Figma".
Use the standard job title for roles where the employer's internal title is unusual. If your official title was "Digital Ninja", write "Digital Marketing Specialist (internal title: Digital Ninja)". This helps both parsers and human readers.
Myths to ignore
"White text on white background for keyword stuffing." Modern parsers may extract the hidden text, and human readers may find it. Either way, it does not help.
"You need exactly the keywords the job description uses." Match the substance, not every word.
"You should submit a DOCX because PDFs cannot be parsed." Both formats parse fine as long as the text is selectable.
Practical checklist
- Single-column layout with standard section headings.
- A widely available font between 10 and 12 points.
- Selectable-text PDF.
- Margins between 12 and 25 millimetres.
- Honest, relevant skills section with eight to fifteen items.
- Match the vocabulary of the job description where it is natural.
- No text inside images.
- No tables used to define the layout.
That is the whole optimisation. There is no further trick.