ATS-friendly CV

What an applicant tracking system actually does

ATS stands for applicant tracking system, the software an employer uses to receive and manage job applications. In text messages ATS also means 'at the same time'; on this page it only means the software. The document you upload is a CV in the UK and Europe and a résumé in the US; the parser treats them the same.

Store, parse, search, sometimes rank

An applicant tracking system does four things with your CV. It stores the file against your application. It parses the file, converting it to plain text and sorting the text into fields: name, contact details, employers, titles, dates, education, skills. It lets recruiters search those fields and the full text. Some systems, or add-ons to them, rank or tag applications by how many terms from the job posting appear. What no system does is decide. A recruiter or hiring manager reads the shortlist and chooses who moves forward.

The systems you will meet

These six are the systems you are most likely to meet at large and mid-sized employers in the US, UK and Europe. The application page usually shows the vendor's name in the web address or the footer.

What breaks parsing

Parsers read a file top to bottom and left to right as a stream of text. Anything that interrupts the stream or hides text from it loses information. The usual causes:

• Two columns. The parser reads across both columns on each line, so a job title on the left is joined to a skill on the right.
• Tables. Cells become one run of text in whatever order the file stores them; dates end up far from the jobs they belong to.
• Text boxes. Many parsers skip floating text boxes entirely. A summary in a box may never be read.
• Headers and footers. Many parsers discard page headers and footers, so contact details placed there never reach the system.
• Icons. A phone icon in place of the word 'Phone' leaves a number with no label; an envelope icon is not an email field.
• Skill bars and rating dots. A bar that shows 'Python 80%' as a graphic contains no text, so the skill is not indexed.
• Images with text. A CV exported as an image, or a scanned PDF without a text layer, is a picture. Nothing is parsed.
• Unusual headings. 'Where I made a difference' is not recognised as experience. Parsers look for Experience, Education and Skills.
• Non-standard fonts. Decorative or subset fonts can map characters incorrectly on extraction, turning 'fi' into a blank or a box.

What helps

The fixes are the mirror image of the list above, and none of them make a CV uglier.

• One column, top to bottom. The reading order is the visual order.
• Standard headings: Experience, Education, Skills, and optionally Summary, Certifications and Languages.
• One date format throughout, such as Mar 2021 to Jun 2024 or 03/2021 to 06/2024, with the word 'to' or a plain hyphen.
• Standard bullets. A plain round bullet extracts as a list; a custom glyph may extract as a stray character.
• A real skills section: named tools and methods as text, not as a graphic, so search finds them.
• PDF with a text layer, or DOCX; send whichever the posting asks for.
• Contact details in the body of the page, not the header.

An applicant tracking system does not detect AI text

No system named on this page runs AI detection on a CV. Workday, Greenhouse, Lever, Taleo, SuccessFactors and iCIMS store, parse and search; whether a sentence was written by a person or a model is not a question they ask. What does happen is that a recruiter reads generic text ('results-driven professional with a proven track record') and sets it aside because it says nothing. The problem with AI-written CVs is not detection but emptiness, and invented numbers that fail at interview. The AI CV builder page covers that.

ATS-friendly CV checklist

Ten items, in the order you meet them on the page.

1. Your name, email, phone and city are in the body of the first page, not in a header or footer.
2. The whole document is one column.
3. There are no tables, text boxes or images anywhere in the file.
4. Headings are Experience, Education and Skills, or close variants such as Work experience.
5. Every job shows employer, title, and start and end dates in the same format.
6. Bullets use a standard bullet character and start with what you did.
7. Every number is one you can defend in an interview.
8. Skills are listed as words, not bars or icons, and each one appears somewhere in your experience.
9. The file is a PDF with a text layer or a DOCX, named with your name and the role.
10. You have run the round-trip test below.

The round-trip test you can do yourself

This takes two minutes and shows you roughly what the parser sees.

1. Open the PDF in any viewer, select all the text (Ctrl+A or Cmd+A) and copy it.
2. Paste it into a plain text editor: Notepad, TextEdit in plain text mode, or a blank email with formatting switched off.
3. Read from the top. Does your name come first, then your contact details, then the sections in the order you intended?
4. Check every job: is the title next to the employer, and are the dates next to both?
5. Look for gaps: a missing summary means a text box was skipped; missing contact details mean a header was dropped.
6. Search the pasted text for three skills from the job posting. If they are not found, they were in a graphic or not on the CV at all.

If the order is wrong or text is missing, fix the source file and export again. A DOCX can be checked the same way by saving it as plain text.

What CitedCV does about each item

CitedCV was built so that this checklist passes by default rather than by effort.

Myths about ATS-friendly CVs

Keyword stuffing works

Repeating a term does not move you up a recruiter's search; a search either matches or it does not. A CV that reads like a list of the posting's terms is set aside by the person who opens it. Use the posting's words where they are true of you, once, in the line where you did the thing.

White text is invisible to the system

White text on a white background is extracted like any other text, so it shows up in the parsed record and on the recruiter's screen. When it is noticed it reads as an attempt to game the system, which is a reason to reject.

Most CVs are rejected by the software before a person sees them

This figure circulates widely, and no vendor named on this page publishes it. Systems store, parse and search; a person decides who is rejected. A CV that parses badly is not rejected by the software. It is hard to find, which has the same effect for a different reason.

PDF is unsafe, always send DOCX

A PDF with a text layer parses in all six systems. The PDF that fails is the one saved as an image, which usually happens with a scan or a design-tool export. If a posting asks for DOCX, send DOCX.

What an ATS-friendly CV cannot do

Passing the parser gets your CV read as you wrote it. It does not get you an interview; that depends on the role, the other applicants and the people reading. It does not hide weak content either: a CV that parses perfectly and says nothing specific is still set aside. CitedCV can make the file parse and the numbers verifiable; it cannot write what you did not provide.

Questions people ask

Sources

1. Greenhouse Supportsupport.greenhouse.io/
2. Lever Help Centerhelp.lever.co/
3. Workdaywww.workday.com/
4. iCIMSwww.icims.com/
5. Harvard Office of Career Services, Create a resume, CV or cover lettercareerservices.fas.harvard.edu/resources/create-a-resume-cv-or-cover-letter/
6. MIT Career Advising and Professional Development, How to write a resumecapd.mit.edu/resources/how-to-write-a-resume/
7. Prospects, How to write a CVwww.prospects.ac.uk/careers-advice/cvs-and-cover-letters/how-to-write-a-cv