How Applicant Tracking Systems Actually Read Your Resume

What an ATS is and what it does

An applicant tracking system is software that employers and recruitment agencies use to manage hiring. When you apply through a company’s careers page, you are usually submitting into one. Large job portals have their own candidate databases that work in similar ways, and many companies connect the two.

Most of what an ATS does has nothing to do with judging you. It stores applications in one place, tracks which stage each candidate is at, schedules interviews, sends emails, keeps notes from interviewers, and produces reports for compliance and hiring metrics. For a recruiter handling dozens of open roles, it replaces an inbox full of attachments and a spreadsheet.

The parts that affect your resume directly are three: parsing (turning your file into structured data), search (letting recruiters find candidates), and screening tools (knockout questions and, in some systems, ranking or matching features). Each works differently, and each is worth understanding on its own.

Parsing: turning your file into fields

When you upload a resume, the system extracts the text and tries to sort it into fields: name, email, phone, location, job titles, employers, dates, education, and skills. That structured version is what populates your candidate profile and what many searches run against. Recruiters can usually also open your original file, but the parsed profile is often what they see first in lists and search results.

Parsing is pattern recognition. The software looks for familiar section headings, date formats, and layouts to decide where one job ends and the next begins. When your resume follows common conventions, it generally does well. When it doesn’t, fields end up blank or jumbled — your employer’s name in the job title box, your dates missing, or your skills merged into your address.

You have probably seen the symptom yourself: an application form that pre-fills from your resume and gets half of it wrong. That form is showing you the parser’s output. If the form gets your work history badly wrong, the recruiter’s view of your profile may be wrong in the same way.

What breaks parsing

Parsers differ, and newer ones handle more layouts than older ones. But the same features cause trouble often enough that it is sensible to avoid them when you’re applying through online systems.

• Tables used for layout. Text in table cells can be read in an unexpected order — across rows instead of down columns — so a job title can end up attached to the wrong employer or dates.
• Text boxes and floating shapes. Word-processor text boxes are sometimes skipped entirely or extracted out of order, which is a problem if your contact details or skills live in one.
• Headers and footers. Some systems ignore the document header and footer areas. Putting your name, phone, and email there is a common way to end up with a profile that has no contact details.
• Images of text. A resume exported as a picture, a scanned page, or a name rendered as a graphic contains no machine-readable text. Unless the system runs character recognition, it sees nothing.
• Icons in place of words. A phone icon next to a number is fine; a phone icon instead of a label, or icons used as skill ratings, may come through as stray symbols or nothing.
• Unusual section headings. “Where I’ve Been” or “My Journey” may not be recognised as work experience. Standard headings — Experience, Education, Skills, Projects, Certifications — are recognised almost everywhere.
• Inconsistent or unusual dates. Mixing “Jan 2023”, “03/2024”, and “Summer ’22” makes it harder to work out job durations. Pick one clear format and use it throughout.
• Unusual fonts or special characters. Decorative fonts and some symbol characters can be extracted as garbage, especially from PDFs that don’t embed text properly.

Columns deserve a separate note. Two-column layouts are common and many modern parsers read them correctly, especially when the columns are created with simple, clean formatting and the PDF contains real text. The risk is reading order: a parser may read straight across both columns line by line, interleaving your sidebar skills with your job descriptions. If you use two columns, keep the side column for short, self-contained items like contact details, skills, and languages, and keep the main story — experience, in order — in one column. When in doubt for a high-stakes online application, a single-column version is the safer choice.

File type matters less than how the file was made. A text-based PDF from a word processor or resume builder is generally fine, and so is a Word document. The real problem is a PDF that is essentially a picture. A quick test: open your PDF and try to select and copy the text. If you can paste it into a plain text editor in a sensible order, a parser can probably read it too. If the employer asks for a specific format, use that.

How recruiters search and filter

Once your application is in, a recruiter rarely reads every resume in arrival order for a busy role. More often they work with views and filters: everyone who applied to this job, sorted by date; candidates who answered the screening questions a certain way; or a keyword search across the applicant pool or the wider database.

Keyword search in these systems is usually much like any other search box. A recruiter types terms — “Salesforce”, “B2B”, “Hyderabad” — sometimes combined with AND, OR, and NOT, and gets a list of profiles that contain them. Some systems also let them filter on parsed fields such as location, years of experience, current title, or education. Job portals popular in India add filters for notice period, current and expected salary, and preferred location, which is why it pays to keep those profile fields accurate.

This has two direct consequences for you. First, if the term the recruiter searches for isn’t in your resume, you won’t appear in that search — no matter how qualified you are. Second, if parsing has misread your resume, filters on fields like job title or experience may exclude you even though the information is technically in your file.

Recruiters also search their database long after you applied. A resume that clearly states your role, skills, and location can surface months later for a different opening.

Keyword matching, honestly explained

Keywords matter, but not in the way the myths suggest. There is no universal score you must beat and no ideal number of repetitions. What matters is that the specific skills, tools, qualifications, and job titles relevant to the role appear in your resume in the words people actually search for.

Some systems include matching or ranking features that compare a resume with the job description and suggest the closest candidates. How they work varies by product and is rarely published in detail, and employers decide whether to use them and how much weight to give them. Older approaches lean on literal term matching; newer ones may understand related terms. Either way, the safest strategy is the same: describe your experience accurately using the vocabulary of the job.

• Use the posting’s terms where they are true for you. If they say “client onboarding” and you wrote “setting up new customers”, switch to their phrasing.
• Include both the abbreviation and full form for common terms at least once: “Applicant Tracking System (ATS)”, “Key Performance Indicator (KPI)”.
• Put keywords in context. A skill that appears in both your skills list and an experience bullet is more convincing to the human who reads it next.
• Don’t stuff. Hidden white text, repeated keyword blocks, or pasting the job description into your resume are noticed by recruiters and can get your application discarded outright.

The auto-rejection myth

You will often read confident claims that most resumes are automatically rejected by ATS software before a person sees them. These figures are rarely traced to any verifiable source, and they misdescribe how most hiring works.

In most setups, rejection is a human decision made inside the system. What feels like an instant automated rejection usually has a more ordinary explanation:

• Knockout questions. If the application asked “Are you authorised to work in this country?” or “Do you have a valid driving licence?” and your answer didn’t meet a requirement the employer set, the system may move you to rejected automatically. That is a rule the employer configured, not the software judging your resume.
• Volume. When a role gets far more applicants than it can interview, recruiters stop reviewing once they have a strong shortlist. Applications that arrive later may never be opened.
• Search-based review. If the recruiter works from keyword searches rather than the full list, a resume that lacks the relevant terms simply never comes up.
• Scheduled rejection emails. Some employers send rejections in batches once a role is filled, which can make a human decision look automated.

So the honest version is this: software rarely throws out a well-matched resume on its own, but a poorly parsed or keyword-thin resume can become effectively invisible. That is a reason to format sensibly and write clearly, not to panic or game the system.

Writing for the software and the human at once

Every resume that gets through to an interview is read by at least one person. Optimising purely for software — dense keyword lists, awkward phrasing — produces a document that reads badly to that person. Fortunately, the two audiences want almost the same things: standard headings, clear job titles, consistent dates, specific skills, and achievements described in plain language.

Put the most important information where both will find it. Your current or most recent job title should match the kind of role you’re applying for where that is accurate. Your summary can name the role and your core skills in one or two sentences. Your experience bullets should mention the tools and methods you used, not just the outcomes.

The free resume builder on this site uses templates with real text headings and produces a text-based PDF identical to the preview, which avoids the image-of-text problem. No tool can guarantee how a particular employer’s system will read a file, though, so the checks below are still worth doing.

ATS-friendly resume checklist

1. Your name, phone, email, and city are in the main body of the page, not in a header, footer, or text box.
2. Section headings use standard words: Summary, Experience, Education, Skills, Projects, Certifications.
3. Experience is listed with job title, employer, location, and dates in a consistent format for every role.
4. No layout tables, text boxes, or images containing important text.
5. Skills are written as text, not shown only as bars, stars, or icons.
6. If you use two columns, the main experience section reads in order and the side column holds only short items.
7. You can select and copy all the text from your PDF and paste it in a sensible order.
8. The key skills and job titles from the posting appear in your resume, in context, where they’re true for you.
9. Common abbreviations are spelled out at least once.
10. Nothing hidden: no white text, no keyword blocks, no pasted job description.
11. The file name is clear, such as firstname-lastname-resume.pdf, and the file type matches what the employer asked for.
12. Screening questions are answered carefully and truthfully.

Frequently asked questions