Skip to content
Blog

What an ATS actually does with your CV, and what it doesn't

Applicant tracking systems are surrounded by myths. Here is what they generally do with a CV, what varies between vendors, and what that means for how you format yours.

By Andrii

7 min read

If you have spent any time reading job search advice, you have probably been told that an ATS will reject your CV before a human sees it unless you use exactly the right keywords and the right font. Some of that is true, some of it is exaggerated, and a lot of it depends on the system and on how the employer has set it up.

This post covers what an applicant tracking system (ATS) generally does, where the common advice goes wrong, and the handful of formatting choices that genuinely matter.

What an ATS is

An ATS is the software a company uses to manage hiring. Job adverts are posted from it, applications land in it, and recruiters and hiring managers use it to review candidates, move them through stages, schedule interviews and send emails. Workday, Greenhouse, Lever, iCIMS, SmartRecruiters, Taleo and many others fall into this category, along with plenty of smaller tools.

For most companies, the ATS is first of all a database and a workflow tool. It exists so that a recruiter handling many open roles does not have to manage applications by email and spreadsheet.

What it does with your CV

When you apply, the details of what happens vary by vendor, but the broad steps are similar.

It stores the file

Your original CV is usually kept as uploaded. When a recruiter opens your application, many systems show that original file, or a preview of it, alongside the extracted data.

It parses the text

Most systems try to read your CV and pull out structured information: name, contact details, job titles, employers, dates, education and skills. This is called parsing. It is what fills in the application form for you, sometimes badly, when you upload a CV.

Parsing is where formatting matters most. The parser has to turn your document into plain text and then guess which part is which. Layouts that look fine to a person can confuse it.

It makes your application searchable

Recruiters can search their candidate database, and often filter by things like location, job title or answers to screening questions. This is where keywords have a real effect: if a recruiter searches for "Terraform" and your CV says "infrastructure as code" but never names the tool, you may not appear in that search.

It may rank or score applications

Some systems, or add-ons to them, compare applications with the job description and produce a match score or ranking. How this works differs a lot between vendors, and employers decide whether to use it and how much weight to give it. Nobody outside a given vendor can tell you exactly how its scoring works, so be wary of anyone who claims to know the precise formula.

What it usually doesn't do

This is where the myths live.

It does not usually "reject" CVs on its own. The most common automatic rejections come from knockout questions the employer set up: "Do you have the right to work in the UK?", "Do you have at least three years of experience with Java?" Answer no to a required question and the system may move you out of the running. That is a rule the employer configured, not the software judging your CV. Whether an employer uses automatic rejection based on CV content at all, and how, varies.

It does not read your CV the way a person does. A parser does not understand that your two years of "looking after the company's AWS estate" means cloud infrastructure experience. It sees words. That is why naming tools and skills plainly helps.

It does not replace the human read. In most hiring processes, a person still reads the CVs that make it through the first stage. Stuffing a CV with keywords can help it surface in a search, and then fail completely when a recruiter actually reads it.

Formatting choices that genuinely help

Most of the formatting advice for ATS comes down to one idea: make it easy for software to extract your text in the right order.

  • Use a single-column layout for the main content. Multi-column layouts can come out of a parser with the columns interleaved, so your job titles end up next to someone else's dates. Some parsers cope well with columns and some do not.
  • Keep text as text. Text inside images, charts or text boxes may not be extracted at all. Skill bars and star ratings carry no information once they are turned into plain text.
  • Use standard section headings. "Experience", "Work experience", "Education", "Skills". A parser looking for work history may not recognise "Where I've been".
  • Put contact details in the body, not only in the header or footer. Some parsers skip headers and footers.
  • Write dates consistently. "Mar 2021 – Jan 2024" or "03/2021 – 01/2024", used the same way throughout, is easier to read for both software and people.
  • Send a PDF with real text, or a Word file. A PDF exported from a word processor or CV tool contains selectable text. A scanned CV is an image, and most parsers will get nothing from it unless they run text recognition. When an employer asks for a specific format, use that.

Fonts, colours and a bit of design generally do no harm, as long as the text underneath is clean. You do not need a plain black-and-white document to get through an ATS.

Keywords: the part that matters, and the part that doesn't

Using the same terms as the job advert helps in two places: recruiter searches and any matching or scoring the employer uses. It also helps the human reader, who is scanning for the same terms.

What helps:

  • Naming tools and technologies explicitly: "Kubernetes", "PostgreSQL", "React", not just "container orchestration" or "a modern front-end framework".
  • Including both the acronym and the full term where it is natural: "continuous integration (CI)".
  • Using the advert's wording for work you actually did.

What doesn't help, and can hurt:

  • Pasting the job description into the CV in white text. Parsers extract it like any other text, so it can show up in the parsed profile or a plain-text view, and it looks like what it is.
  • Repeating the same keyword many times. It reads badly to people, and there is no reliable evidence that repetition helps across systems.
  • Listing tools you have not used. It may get you into a search result, and then into an interview question you cannot answer.

How to check your own CV

The simplest test is to see what your CV looks like as plain text. Copy everything from your PDF and paste it into a plain text editor. If sections come out in the wrong order, text is missing, or your name and email have vanished, a parser may have the same problem.

Our free ATS checker does a version of this for you: upload a PDF or Word file and it shows the text that can be extracted from it, with word and line counts, so you can spot missing or jumbled content. It does not claim to predict how a particular employer's system will score you, because no tool honestly can.

The short version

  • An ATS is mostly a database and workflow tool, and people still read CVs.
  • Parsing is where formatting matters: single column, real text, standard headings.
  • Keywords help your CV get found and read, when they describe experience you really have.
  • Anything that games the system tends to backfire once a person reads the CV.

If you want a CV that uses the advert's language and that you can talk through in an interview, that is what MyCVStack is built for. See how it works.