What an ATS resume score measures, and what number is enough

If you’ve used any resume checker, you’ve seen the number: your resume scores 41, or 68, or 89 against a job posting. It feels like a grade. It isn’t, and understanding what an ATS resume score actually measures is the difference between using it well and chasing a meaningless 100.

What the score actually is

An ATS resume score is, at its core, coverage: how much of what the job posting asks for can be found in your resume’s text. A posting mentions “funnel optimization”, “growth experiments”, “stakeholder management”, “product analytics”. A scorer checks which of those your resume contains, weights them, and returns a percentage.

That’s it. It is not an assessment of whether you’re good at your job, how impressive your career is, or how nice your formatting looks. A brilliant engineer’s resume can score 30 against a posting written in different vocabulary; a mediocre one can score high by parroting it.

Sarto’s score works exactly this way, deliberately: it’s computed deterministically from the posting’s own wording, so the same resume and posting always produce the same number. No mystery meter, no “AI vibes”: you can see the exact missing keywords behind every point.

A score is a coverage report, not a grade: 41 means the resume answers 41% of what this posting screens for, and the chips are the gap.

What ATS software really does with it

A common myth says applicant tracking systems auto-reject anyone under some threshold. The reality across the industry is less dramatic but just as consequential: most systems rank, filter, and surface candidates rather than auto-rejecting them. Recruiters search the applicant pool by the posting’s terms, sort by relevance, and read from the top.

The practical effect is the same either way: if your resume doesn’t speak the posting’s language, it sits where nobody looks. You don’t get a rejection explaining that. You get silence. (If that’s where you are right now, start with why applications get no response.)

So what number is enough?

There is no universal pass mark, and any tool that claims one is guessing. But the useful reading of the number is comparative:

• Low coverage (roughly, under 50): your resume is invisible for this role’s vocabulary. The posting and your resume are describing overlapping work in non-overlapping words. Worth fixing before sending.
• Mid coverage (50–75): you’ll register in searches for some requirements and miss others. Look at which keywords are missing: if the missing ones are core to the role, close them; if they’re peripheral, you may be fine.
• High coverage (75+): your resume answers the posting in its own terms. Further points come mostly from keywords you’d have to stretch or fake, so don’t.

The target is not 100. A resume that mirrors every phrase of a posting reads as engineered, and the last points usually require claiming things you haven’t done. The target is: no core requirement you genuinely have goes unstated.

How to raise the score honestly

The only legitimate way to raise coverage is to close vocabulary gaps on experience you actually have:

1. Extract the posting’s keywords (here’s how to find them).
2. For each missing one, ask whether you’ve done it. If yes, rephrase an existing bullet into the posting’s language. If no, leave it missing.
3. Re-score and repeat until the remaining gaps are things you genuinely lack.

That workflow (extract, map honestly, rephrase, re-score) is the whole method behind tailoring your resume to a job description.

A real example of the honest gap-close

The gap-close, change by change: reorder and rephrase only, nothing invented.

In Sarto’s demo data, a product manager’s resume scores 41 against a growth-PM posting. The resume says “worked with marketing and support teams to prioritize the roadmap”; the posting wants “stakeholder management”. It says “cut onboarding drop-off by 22% through funnel experiments”; the posting wants “funnel optimization” and “growth experiments”. After rephrasing those existing facts into the posting’s vocabulary (nothing added, nothing invented), the same resume scores 89. The candidate didn’t change. The resume finally described them in the words this employer was searching for.

Where Sarto fits

Sarto shows you the score and the missing keywords before you pay anything, then does the rephrasing pass for you: every change is shown as a diff you can reject, and the server validates that nothing was invented: it can only reorder and rephrase what your resume already contains. Import your resume once, paste a posting, and the 41-to-89 pass takes about a minute instead of an evening.