What Is an ATS? How Applicant Tracking Systems Actually Work (2026)

Employers use different combinations of parsing, application fields, recruiter search, screening questions and optional matching. A public resume checker cannot determine your employer’s settings. This guide separates those functions so you can inspect your document without treating a diagnostic score as a hiring prediction.

What an ATS is

An applicant tracking system is a database application that companies use to receive, organize, and manage job applications at scale. Think of it less as a robot reader and more as a structured filing cabinet: it ingests your resume, tries to convert it into structured data (name, contact, work history, skills, education), stores that data, and makes it searchable by recruiters.

The major platforms in wide use include Workday, Greenhouse, Lever, iCIMS, Taleo, and BambooHR. Each has its own parser, its own search interface, and its own quirks — which is part of why "ATS optimization" advice is never one-size-fits-all.

What an ATS actually does (the four steps)

1. Parse your resume into structured fields

When you submit, the system runs your file through a parser that attempts to extract structured data: your name and contact info, each job title, employer, and date range, your education, and a list of skills and keywords. The quality of this extraction varies significantly by system and by how cleanly your resume is formatted. Parsing errors — a garbled job title, a missing date, skills lumped into the wrong field — are invisible to you but make your record harder for recruiters to find.

Workday documents that its parser populates fields from a resume and that results vary with format and word order. This is a platform-specific example, not a rule for every ATS. Read Workday's resume-parsing documentation.

2. Store the application in an ATS database

An ATS can connect a candidate profile with job-specific applications and recruiting workflow details. For example, Greenhouse describes its candidate profile as centralizing candidate and application information, documents, interview stages, notes, and tasks. That is one product's workflow, not a standard record layout for every ATS. Read the Greenhouse candidate profile documentation.

3. Support recruiter keyword search and filtering

This is the step most directly affected by your word choices. In a literal-search workflow, a recruiter might combine terms such as "product manager", "SQL", and "B2B SaaS". A resume that uses different terminology—or whose relevant text failed to parse—may not appear in that result. Other systems add semantic or criteria-based matching, so exact wording matters without being the only mechanism in the market.

4. Optionally rank or score candidates

Some systems calculate a match or fit score; others do not produce a numeric resume score at all. A score may help a recruiter sort or shortlist candidates. Automated actions, where an employer enables them, depend on that product's configuration, screening questions, or explicit criteria—not on one universal public pass mark.

What you should not assume an ATS does

• Do not assume it reads like a person. Parsing starts by extracting text and fields. Some newer matching systems can connect related skills, but none can verify the quality of your work from a phrase alone.
• Do not assume one score controls rejection. Some systems have no match score. Automatic decisions, where configured, may instead depend on knockout questions, explicit employer criteria, assessments, or workflow rules.
• Do not confuse matching with proof. Software may recognize a skill or qualification, but a human still judges whether the surrounding achievement is relevant, credible, and strong enough for the role.
• Do not treat “the ATS” as one model. Workday, Greenhouse, Workable, Lever, and other platforms expose different combinations of parsing, search, ranking, questions, and automation—and each employer configures its own workflow.

What the common ATS advice gets wrong

Four claims cause most of the confusion: that ATS rejects exactly 75% of resumes, every system assigns a score, PDFs are unreadable, and one percentage is a universal pass mark. Each claim turns a product-specific risk into an industry rule. We traced the original 75% source and checked all ten major claims against current vendor documentation in our primary-source ATS myths fact-check.

Why qualified people still get filtered out

Two possible issues to inspect are extraction errors and terminology differences. A resume built with tables, columns, headers/footers, or unusual fonts may parse poorly — your job titles end up in the wrong field, or your skills section is dropped entirely. In a literal-search workflow, a resume may not surface if the recruiter searches for "Agile project management" and your resume only says "Scrum" and "sprint planning".

The fix is a clean, single-column format with standard section headings — and careful mirroring of the job description's own language in your skills and experience sections.

Related: how an ATS reads your resume (parsing explained) · ATS-friendly resume format · why resumes get rejected by ATS