You run your résumé through a scanner and it hands back "72% ATS match." A different tool gives the same document 91%. A third one doesn't score it at all, just lists warnings. None of them agree, and none of them tell you what a recruiter at the company you're applying to would actually see. That's not a bug in any one tool — it's because "ATS score" isn't a standard, universal thing to begin with.

What these scoring tools are actually doing

Most free résumé scanners work the same way under the hood: they parse your document into text, compare that text against the job description you paste in (or against a generic keyword list if you don't), and produce a percentage based on how much overlap they find. Some also run basic formatting checks — single column or not, standard section headings or not — and fold that into the same number.

That's a genuinely useful diagnostic. Keyword overlap with a real posting is a reasonable proxy for relevance, and a formatting check can catch real parsing problems before you submit anywhere. What it isn't is a preview of a hiring decision, or even a preview of what the employer's actual system shows.

What a real ATS shows a recruiter

The systems companies actually use — Workday, Greenhouse, iCIMS, Taleo, SAP SuccessFactors, and the rest — don't typically hand a recruiter a single public-facing percentage the way consumer scanning tools do. What a recruiter sees is closer to a structured record (your parsed name, employers, dates, skills) and, in systems with matching features turned on, a sorted list of candidates ranked against the requisition. Some newer platforms do surface an internal match score to recruiters, but it's rarely the headline number the job seeker ever sees, and it isn't computed the same way two different consumer tools compute theirs. There's no one formula being reverse-engineered here — every ATS parses and ranks a little differently, which is exactly why a single "your score is 84%" claim from a free tool can't be a precise prediction of anything.

So is the score pointless?

Not pointless — just easy to over-trust. Treat a score the way you'd treat a spell-checker's underline: a signal worth investigating, not a grade to optimize toward. If a tool flags that your résumé is missing terms the posting uses repeatedly, that's worth a look — maybe you do have that experience and just phrased it differently. If a tool flags a formatting issue (a table, a text box, an icon standing in for a skill), that's worth fixing regardless of what percentage it produces, because parsing failures are a real, well-documented problem independent of any scoring gimmick.

What's not worth doing is chasing the number itself — rewriting a résumé sentence by sentence to nudge a free tool's percentage from 78% to 95%. Two different scanners will disagree with each other, and neither one is the system the employer is actually running.

What to check instead

  • Does it parse cleanly? Single-column layout, standard section headings, contact info in the body rather than a header or graphic — the structural basics that let any parser read you correctly.
  • Does it actually match the posting's language? Not keyword-stuffing — genuinely reflecting the skills and terms that specific job description uses, where you truthfully have that experience.
  • Would a human reading it for thirty seconds understand what you do? A résumé that scores well on a keyword tool but reads as a jumble to a person hasn't actually solved the problem.

If you want a straightforward check on the structural side without a mystery percentage attached, DraftCareer's free ATS compatibility checker scans for the specific formatting issues that cause real parsing failures, rather than producing a score built on an unknown formula.