What an ATS score actually measures, what counts as a good resume ATS score, and why the "ATS auto-rejects 75% of resumes" claim is a myth. Includes real score bands and a free checker.
You ran your CV through an ATS score checker and got a 62. Is that bad? Should you rewrite the whole thing, or keep applying and see what happens?
The answer depends on what's actually dragging the number down, and on understanding what "ATS score" means in the first place. There's a lot of confident-sounding advice online about "beating the bots," and a fair amount of it is wrong.
This guide explains what an ATS score actually measures, what counts as a good resume ATS score, and clears up the single biggest myth in the whole topic: the claim that Applicant Tracking Systems auto-reject three out of four CVs before a human ever opens them. They don't work that way, and believing they do leads people to fix the wrong things.
What Is an ATS Score?
"ATS score" gets used to mean two different things, and mixing them up is where most of the confusion starts.
A job-specific match score. Tools like Jobscan compare your CV against one specific job description and return a percentage based on keyword and phrase overlap. This score changes every time you check it against a different posting, because it's measuring fit to that posting, not the CV's overall quality.
A general CV-quality score. Other tools, including our own free ATS resume checker, score the document itself out of 100: is it structured well, can software parse it cleanly, do the bullets show evidence rather than vague duties. This kind of score doesn't require a job description at all, because it's assessing the CV as a standalone document.
Neither of these is the number a real employer's software shows a recruiter, because it doesn't show one to you. Platforms like Workday, Greenhouse, Lever, and iCIMS don't publish a score back to the applicant. Every "ATS score" you've ever seen came from a third-party checker sitting outside the actual recruiting software, estimating how a parser and a recruiter's filters would likely treat your CV.
That distinction matters because it changes what a "good score" even means. We'll come back to concrete numbers below, but first, the myth that shapes how most people think about this entire topic.
The "ATS Auto-Rejects 75% of Resumes" Myth
If you've read anything about ATS in the last few years, you've probably seen a version of this claim: 75% of resumes are rejected by ATS software before a human ever sees them.
It's repeated constantly. It's also traceable to a single, unsupported source.
According to a breakdown by The Interview Guys, the figure originated from a 2012 marketing pitch by a company called Preptel, which sold resume optimisation services and shut down the following year. No study, survey, or methodology was ever published to back the number up. It was a sales claim, and it outlived the company that made it by more than a decade.
What does the evidence actually show? The same breakdown cites recruiter estimates in the 90 to 95% range for how many applications get at least a human look, not the 25% the myth implies. Real ATS platforms don't have a built-in "reject 75% of applicants" function. What they do have is filtering and sorting based on rules that a human recruiter configures, things like required work authorisation, minimum years of experience, or specific certifications.
There's a more nuanced, better-sourced data point worth knowing here too. Harvard Business School's 2021 Hidden Workers report found that most employers surveyed admitted their own automated hiring systems filter out qualified candidates, and that a resume gap of six months or longer is, on its own, enough to trigger exclusion at a large share of them. That's a real and well-documented problem. But notice what it actually describes: a human decided that a six-month gap should disqualify a candidate, and configured the software to apply that rule consistently. The software didn't decide anything on its own; it executed a filter a person built.
None of this means ATS software is irrelevant. Adoption is genuinely close to universal: Jobscan's analysis of Fortune 500 companies found 97.8% use some form of ATS. What's mythical isn't the software, it's the idea that it silently auto-bins most qualified applicants with no human ever reviewing the outcome.
The practical takeaway: you're not writing your CV to "trick a robot" into letting you past a gate it was never built to slam shut. You're writing it so that (1) the software extracts your information correctly, and (2) it ranks well against the specific, human-set criteria for the role. That reframing changes which fixes are actually worth your time, and it's the whole reason ATS scoring tools exist.
What ATS Software Actually Does
Since the auto-reject framing is wrong, here's the more accurate, three-step version of what happens after you click "Apply":
Parsing. The system extracts your name, contact details, job titles, dates, and skills into structured fields. This is where formatting problems bite: multi-column layouts, text in tables, and contact details hidden in a header can all cause information to be misread or dropped entirely.
Matching and ranking. The software compares extracted fields against the job's requirements, most commonly skills, job title, education, and years of experience, and either filters or sorts candidates by fit.
Human review. A recruiter opens the ranked or filtered list and makes the actual decision. This is the step that was always human, myth or no myth.
For the full breakdown of how parsing, keyword matching, and ranking work across specific platforms, see our ATS-friendly resume guide. The rest of this post focuses on the part that's specific to scoring: what a checker is actually measuring when it hands you a number, and what counts as good.
What Is a Good ATS Score?
There's no single universal answer, because "score" means different things depending on the tool, as covered above. But here are real, concrete benchmarks rather than a vague "aim high."
For a job-specific match score (comparing your CV to one posting): Jobscan's own guidance is that a score of 75% or higher means you're a good match and likely to get an interview. This kind of score is meant to move, it should go up each time you tailor your CV to a new posting.
For a general CV-quality score (assessing the document on its own): our free ATS resume checker scores out of 100 using four fixed bands. The full method is public on our scoring methodology page, but here's the summary:
Score
Band
What it means
85-100
Excellent
Clean for parsing and substantively strong. Competitive as it stands
70-84
Strong
Solid overall. A handful of targeted fixes moves it into the top band
50-69
Fair
Readable, but underselling you. There are clear, specific things to fix
0-49
Needs work
Something structural or substantive is holding the CV back
A score in the 70s is a genuinely fine place to be. It doesn't mean stop improving, but it doesn't mean panic either. A score under 50 usually points to something fixable and specific, like a two-column layout a parser can't read cleanly, or a CV made up almost entirely of duty statements with no quantified outcomes, rather than a vague sense that the whole document is wrong.
One caveat, straight from our own methodology: no tool, ours included, can tell you the exact score a specific employer's proprietary system would assign. Every company configures its own rules and weightings, and most never expose a number to anyone outside the company. What a good checker can tell you is whether your CV is cleanly machine-readable and substantively strong, which helps with every ATS and every human reader afterwards. That's the part you can actually control, and it's worth checking for free before you send another application into the void.
Why a 0 to 100 breakdown rather than a simple pass or fail? Because "pass or fail" hides exactly the information you need. A CV that scores 68 because it's missing quantified results in the experience section needs a completely different fix from one that scores 68 because two-column formatting is scrambling the parsing. A single band tells you roughly where you stand; the line-by-line breakdown underneath it tells you what to actually change. That's also why a trustworthy checker should be consistent: the same CV, checked twice without edits, should return the same score both times. If a tool gives you a different number on every run with no changes made, treat the result as noise rather than signal.
What ATS Scoring Tools Actually Measure
Whichever tool you use, a legitimate ATS score checker is almost always assessing some version of the same four things.
Parseability and Formatting
Can the software extract your information at all? This checks for a single-column layout, standard section headings the parser will recognise ("Work Experience," not "Career Journey"), no content trapped in tables, text boxes, headers, or footers, and no contact details hidden behind icons that many parsers simply skip. This is the most mechanical part of a score, and also the easiest to fix once you know what's wrong. Our ATS-friendly resume guide covers the full formatting rulebook in detail.
Keyword and Skills Match
Does your CV contain the specific terms a recruiter's filters or a human reader would search for? This includes exact phrases from the job description, natural variations, and both the acronym and the full term for anything abbreviated ("Search Engine Optimisation (SEO)"). See our guide to resume keywords for how to find and place them, and how to tailor your resume to a job description for doing this properly per application rather than once, generically.
Quantification and Impact
Do your bullet points show outcomes, or just list duties? A line like "Responsible for social media" reads as filler. A line with a number, a scale, or a result reads as evidence.
Weak (duty-only):
Responsible for managing the customer support team and improving response times.
Stronger (quantified):
Managed a 6-person customer support team, cutting average first-response time from 4 hours to 45 minutes across 1,200+ monthly tickets.
Both describe the same job. Only one gives a scoring tool, and a human reader, something concrete to credit you for. Our guide on quantifying your achievements walks through how to find numbers even in roles that don't feel naturally metric-driven.
Completeness and Structure
Does the CV have the sections a reader expects, in a sensible order, with nothing thin or missing for the level of role you're targeting? A one-line "Skills" section under a senior engineering title, or an experience section with no dates, will cost you here even if every individual sentence is well written.
How to Actually Improve Your ATS Score
In rough order of impact:
Get a real breakdown first. Don't guess which of the four areas above is weak. Run your CV through the free ATS resume checker, which is exactly what it's built for, and see which specific lines and sections are costing you points.
Fix structural issues before anything else. A parsing problem (a two-column layout, contact info in a header) can suppress your entire CV regardless of how good the content is. Fix formatting first.
Tailor per application. A generic CV can't score well against every job at once. Tailoring your resume to the job description is the single highest-leverage habit for the keyword and match portion of any score.
Add real keywords, not stuffed ones. Weave in the specific terms from the posting naturally. Our resume keywords guide covers placement without sounding robotic.
Quantify what you can. Go bullet by bullet and ask "compared to what, and by how much." Even rough estimates beat no number at all.
Re-check after editing. Scores are meant to move. If a fix doesn't change the number, you may be solving the wrong problem.
This is also where a tool like JobSprout's AI assistant can help mid-edit, suggesting a stronger, more specific way to phrase a bullet you've flagged as weak, without you needing to start from a blank page.
Frequently Asked Questions
Is a low ATS score the same as being rejected?
No. A low score from a third-party checker means the tool has flagged specific weaknesses, usually in formatting, keyword match, quantification, or structure. It's a diagnostic, not a verdict, and as covered above, real ATS software doesn't auto-reject the way the popular myth suggests. A low score is a to-do list, not a rejection letter.
What's a good ATS score out of 100?
Using bands like ours, 70 and above is solid (Strong or Excellent), 50 to 69 is workable but has clear, fixable gaps, and under 50 usually points to something structural, like a parsing problem or a lack of quantified achievements, that's worth addressing before you send out more applications.
Do real ATS platforms show me a score?
No. Employer-facing systems like Workday, Greenhouse, and Lever don't return a number to the candidate. Every "ATS score" you've seen comes from a separate checker tool estimating how the CV would likely be parsed and ranked.
Can I get a 100 on an ATS score checker?
It's possible but not the goal to chase for its own sake. A CV that's perfectly formatted, has every keyword, and is fully quantified will usually already score in the Excellent band. Chasing the last few points by over-stuffing keywords tends to hurt readability for the human who reads it after the software does its part, and most checkers, including ours, are built to catch that pattern rather than reward it.
Does a high ATS score guarantee an interview?
No. A high score means your CV is cleanly readable by software and substantively strong on its own terms. It doesn't account for how many other applicants applied, internal referrals, or a hiring manager's specific preferences. Think of it as removing a controllable obstacle, not as a guarantee.
How is a general CV-quality score different from Jobscan's match rate?
A match rate compares your CV to one specific job posting and changes every time you check it against a different one. A general quality score, like the one from our free checker, evaluates the document on its own: parseability, keyword and skills presence, quantification, and completeness, without needing a job ad at all. Both are useful for different questions: one asks "how well does this fit that job," the other asks "is this a well-built CV."
Should I trust every free ATS checker?
Be a little sceptical of any tool that gives you a precise percentage claiming to predict how a named employer's specific system will treat your CV. As covered above, no outside tool has access to a company's private configuration, so that kind of precision is a guess dressed up as data. A trustworthy checker will tell you what it actually measures, publish its scoring bands or methodology openly, and return a consistent score for an unedited CV rather than a different number every time you refresh the page.
See Your Actual Score
Reading about ATS scores only gets you so far. The fastest way to know where your CV actually stands is to check it.
Run your CV through the free ATS resume checker, no signup required, and get a full breakdown across formatting, keyword match, quantification, and structure in seconds. Curious exactly how the number is calculated? The scoring methodology is fully public, nothing held back.
