An applicant tracking system parses your resume into structured fields and stores it in a database that recruiters search, filter and sometimes rank. “Keyword matching” is mostly that search: if your resume does not contain the terms a recruiter or a configured filter looks for, it does not come up. There is no universal score, and employers configure these systems very differently.
What an ATS is
An applicant tracking system is the software an employer uses to post jobs, collect applications, and manage candidates through a hiring process. Well-known products include Workday, Greenhouse, Lever, iCIMS, Taleo and SuccessFactors, and they differ substantially in features and defaults. What they share is that every application becomes a record in a database, and the resume becomes text and fields inside that record.
Step 1: Parsing
When you upload a resume, a parser extracts its text and tries to identify contact details, employment history (employer, title, dates), education, and skills. Parsers are reasonably good with conventional layouts and worse with multi-column designs, text inside images or text boxes, decorative section headings, and content in headers and footers. A field the parser misses is empty in the record, and empty fields do not match anything.
This is why formatting advice for these systems is so consistent: one column, real text, standard headings, consistent dates. It is not about tricking anything; it is about the extraction working. How to pass an ATS has the checklist.
Step 2: Search and filters
This is where most “keyword matching” happens. Recruiters build shortlists by searching the database: a job title, a skill, a certification, a location, a years-of-experience figure, sometimes boolean combinations. Employers can also configure filters or screening questions that automatically tag or exclude applicants on specific fields.
The consequence is simple and often misunderstood. If a recruiter searches for “supplier management” and your resume says “coordinated vendor deliverables,” you are not in the results, no matter how well qualified you are. This is not a judgement about you. It is a text search that did not match. Some systems apply synonym expansion or semantic matching; many do not, and you cannot know which from outside.
Step 3: Ranking and scoring, where it exists
Some systems offer ranking or “match” features that compare a resume to a job description and produce a score, and some employers use them to sort the applicant list. Other employers have the feature and ignore it, and others do not have it. Where it exists, the inputs are the same parsed fields and the same text: terms from the requisition, years, titles, education. There is no industry-wide algorithm and no single threshold, and any tool that shows you “your ATS score” is showing you its own estimate, not the employer’s.
What this means for how you write
- Use the posting’s vocabulary for real experience. The search terms a recruiter uses come from the requisition. If you did the thing, name it the way the posting names it.
- Put primary terms where they will be parsed and read. In the experience bullets and the summary, not only in a skills list, and not in a header or image.
- Do not stuff. A skills line with forty terms may match a search and then fail the human who opens the record. Terms attached to a real bullet do both jobs.
- Do not add terms for experience you lack. A match that leads to an interview you cannot support is worse than no match.
- Make the fields parseable. Dates, titles and employers in a conventional structure.
Checking a resume against a posting
The useful check is a comparison: the posting’s concrete requirements on one side, the evidence in your resume (and beyond it) on the other, with the vocabulary differences called out. That is what FilterProof’s ATS resume checker and resume keyword optimizer do against your own evidence. FilterProof reports what it found — matched, described in other words, missing — and does not present its result as the employer’s score, because it cannot know that.
Frequently asked questions
Do all applicant tracking systems score resumes?
No. Some offer a ranking or match feature; some employers use it, some ignore it, and some systems do not have one. Recruiter search and configured filters are the more universal mechanism, and they are text matching on parsed fields.
Does the ATS read PDF or Word better?
Modern parsers handle both when the file contains real text. Problems come from layout — columns, text boxes, images — not from the container format. If the posting or form specifies a format, use that one.
Is an “ATS score” from a resume tool the same as the employer’s?
No. Any score from an external tool, including FilterProof’s alignment analysis, is that tool’s comparison of your document to the posting. The employer’s system, if it scores at all, uses its own configuration. Use external checks to find vocabulary and evidence gaps, not to predict an outcome.