An ATS score is a number that claims your resume is a good match for a job. Most tools that produce one will not tell you how. This page does: the two scores Resumedit computes, which one you are looking at where, the exact weight each component carries, and what the resulting number does and does not mean.
The short version
There are two. The big number on your analysis result is a quick match score built mostly from keyword coverage. The five-component detailed score, weighted by role, is the one on the resume review screen and the one most of this page describes. They can disagree, and when they do it is because they are measuring different things.
Which number am I looking at?
This matters more than anything else on this page, because the two scores weigh different things and a change that moves one may not move the other.
Quick match score — the ring on the analysis result
When you analyse a resume against a posting, the large score in the ring is this one. It has five categories and none of them are recency or achievements:
- Hard skills — 55% to start, and it grows. If the posting names no qualification, education's share moves here, taking it to 65%. When validated keyword extraction is available it gains a further 15 points, so it can reach 80%
- Job title — 15%, falling to 10% when hard skills takes its boost
- Soft skills — 10%, falling to 5%
- Education — 10%, or nothing when the posting asks for no qualification
- Other keywords — 10%, falling to 5%
This is a keyword-coverage score with some context around it. It is deliberately close to what the well-known keyword scanners report, which is why it is the headline: it is the number you can compare against the tool your recruiter friend told you to use. It does not look at whether your bullets describe outcomes, and it does not look at whether your relevant experience is recent.
Detailed score — the breakdown on the resume review screen
Five components, weighted by the role in the posting, capped so keywords cannot dominate. This is the score the rest of this page explains, and it is the more useful of the two for deciding what to actually change, because achievements and recency are in it. If you want to know why a resume is weak rather than how many terms it is missing, read this one.
Why the headline number moves when you add keywords and little else does
Because the headline score is mostly keyword coverage, adding genuine terms from the posting moves it a lot. Rewriting a duty into an outcome moves it not at all, and moves the detailed score considerably. Neither is wrong. Use the headline to check coverage and the breakdown to decide what to write.
First, what an ATS actually does
An applicant tracking system is a database with a parser attached. When you upload a resume, the parser extracts text and tries to map it into structured fields: contact details, employment history with dates, education, skills. A recruiter then searches and filters that database.
Two things follow from that, and they are the source of most of the confusion around the subject.
The first is that an ATS does not usually reject anyone. It is a filing system. What rejects you is a recruiter running a search your record does not match, or never running one that would have found you. The distinction matters because it changes what you should optimise for: not appeasing an algorithm, but being findable and legible in a database.
The second is that no ATS publishes a score. There is no number inside Workday or Greenhouse that a candidate can raise. Any ATS score you are shown, including this one, is a model of how well your resume matches a specific job, built by whoever is showing it to you. That is worth stating plainly, because a score presented as if it came from the employer's system is a fiction.
What the Resumedit score measures
Seven components, each scored from 0 to 100. The first five compare your resume against the job description you provide, and switch off without one. The last two, section completeness and bullet quality, read your resume on its own and are scored either way, because a resume with no summary or no bullets that carry a number is a real weakness whatever job it is measured against.
Keywords Match
How many of the skills, tools and domain terms the job description asks for appear in your resume. Matching is lenient about form: a term is credited whether it appears as written, as a plural, or as a close morphological variant, so "managed migrations" counts for "migration". It is not lenient about absence. If the posting names a tool you have never written down, that term is missing.
Title/Seniority Alignment
How close your job titles are to the one being filled, in both function and level. A senior backend engineer applying to a staff backend engineer role scores well here. The same person applying to an engineering manager role scores lower, correctly: it is a different job, and the resume has to argue for the move rather than assume it.
Recent Experience
Whether the relevant experience is current and whether the dates are legible at all. A resume with unparseable or missing dates scores lower here than one with a clean, consistent range, because a reader cannot tell whether the relevant work was last year or nine years ago.
Achievements Quality
Whether your bullets describe outcomes or duties. "Responsible for the deployment pipeline" is a duty. "Cut deployment time from 40 minutes to 6" is an outcome. This is the component most resumes have the most room to move on, and the only one that improves purely by rewriting what is already there.
Education/Certs
Whether the degrees, certifications and licences the posting asks for are present and findable. Small by default, and deliberately so, but it is the component that swings hardest by role: it is worth five times as much on an entry-level profile as on a default one.
Section Completeness
Whether your resume carries the four sections a reader expects: a summary, experience, education and skills. Contact details are not counted here, because a separate formatting check already looks for them. Scored with or without a job description, because a missing section is a gap whatever role you are applying to.
Bullet Quality
How your experience bullets are written, independently of Achievements Quality above. A bullet loses the point here when it opens on "Responsible for", "Worked on" or "I", or when it is too short to say anything or long enough to read as a paragraph. Also scored with or without a job description.
The last two are newer than the first five, which is why they are easy to miss in older explanations of this score, including earlier versions of this page. They each carry a flat 10% before the five job-relative components above are scaled down to make room, which is why the published weights below do not sum to 100% on their own.
The weights, published
These are the actual profile weights the analyser uses for the five job-relative components, before section completeness and bullet quality join at a flat 10% each and the table is scaled back down to fit them. Read the numbers below as relative weight against each other, not as the final percentages of the score.
| Component | Default | Entry level | Executive |
|---|---|---|---|
| Keywords Match | 40% | 35% | 25% |
| Title/Seniority Alignment | 25% | 10% | 35% |
| Recent Experience | 15% | 10% | 10% |
| Achievements Quality | 15% | 20% | 25% |
| Education/Certs | 5% | 25% | 5% |
On the default profile, blending section completeness and bullet quality in at 10% each and scaling the rest down to fit gives effective weights of roughly: Keywords Match 33%, Title Alignment 21%, Recent Experience 13%, Achievements Quality 13%, Education/Certs 4%, Section Completeness 8%, Bullet Quality 8%. Those seven then carry 85% of the final score; a separate formatting check, covered below, carries the other 15%.
Read the entry-level column against the default one and the reasoning is visible. A graduate has no title history to align and little recency to speak of, so those weights fall; the degree and the coursework are the strongest evidence available, so education rises from 5% to 25%. An executive is the mirror image: level and scope are most of the question, so title alignment rises to 35%, and nobody is checking the degree.
How the weights are chosen
There is a profile per role family, and the analyser picks one by reading the job title and description you paste in. A posting for a site reliability engineer resolves to a different weighting than one for a paralegal, because the two are not judged on the same things.
Seniority is carried inside the profile rather than applied on top of it. There is no separate multiplier that reads "senior" or "director" out of the posting and adjusts the weights again; two of the profiles are themselves seniority profiles, and the rest encode a typical level for their role family.
- Entry level and new graduate postings resolve to a profile that puts education at 25% and title alignment at only 10%, which is the right shape when there is not much history to align
- Executive and C-level postings resolve to one that puts title alignment at 35% and keyword match at 25%
The pattern across those two is still the most useful thing to take away: the more senior the role, the less keyword matching carries and the more your titles and your outcomes do. Stuffing an executive resume with tooling keywords is optimising a component worth 25% against one worth 35%.
The keyword cap, and why it exists
Once the profile weights are set, keyword match is capped. It cannot exceed 35% of the total score, or 40% for role families where tooling genuinely is most of the job. Any weight above the cap is redistributed across title alignment, recent experience and achievement quality.
It is never redistributed to education. Education is a small, largely binary component, and pushing weight into it would let a degree compensate for a resume that does not match the job.
The cap exists because a scoring model that lets keyword match run to 60 or 70% is trivially gameable and actively harmful: it rewards pasting the job description into white text at the bottom of the page, and it tells a strong candidate with different vocabulary that they are a bad match. Capping it means a resume cannot score well on keywords alone, and cannot be sunk by keywords alone either.
What the bands mean
Strong match on the components that carry weight for this role. Further keyword work has little left to give; spend the effort on the cover letter and the application itself.
Competitive. Usually one component is dragging: open the breakdown and look for the lowest bar rather than editing everywhere at once.
Worth applying, worth editing first. At this level there is normally a specific, fixable gap: missing terms the posting names, or bullets written as duties.
Either the resume needs real work for this posting, or it is the wrong posting. Both are useful answers, and the second is worth taking seriously rather than editing around.
Each component gets its own band on the same thresholds, which is why the breakdown is more useful than the headline number. A 68 made of five middling components and a 68 made of four strong components and one broken one need completely different edits.
Worked example
Rather than a hand-typed example that can quietly drift from the code as the model changes, our published methodology page runs the live scorer at render time on a fixed resume and a fixed posting, and prints exactly what comes back: all seven categories, the weight each one actually carried, and the resulting score. If your own score ever disagrees with what the two pages together describe, that is a bug worth reporting, not a nuance.
You can also see the model run on real resumes against real postings, one per role, on our ATS resume examples page, each linking through to its own full breakdown.
What happens without a job description
Two of the five components are questions about a specific posting: keyword match and title alignment both need something to match against. When you score a resume with no job description, those components are inactive and their weight is redistributed across the components that can still be measured. The number you get back is a general readability and quality score, not a match score, and it should not be compared with a score produced against a real posting.
What the score deliberately does not measure
- Whether you will get the job. It measures the match between two documents. It knows nothing about the other applicants, the hiring manager or the internal candidate.
- Visual design. A resume can be beautiful and score badly, or plain and score well. Parseability is checked separately from the five components; see the note on that below.
- Truthfulness. Adding a skill you do not have raises the keyword score and loses you the interview. The model cannot tell the difference; the interviewer can.
- Whether the posting is worth applying to. A high score against a badly written job description is a high score against a badly written job description.
Formatting is scored separately, then blended in
Whether an ATS can read your file at all — columns, tables, header regions, graphics, date formats — is checked independently of the seven components above, and every parse issue found is reported as its own flag. That formatting score is not just diagnostic: it is blended into your final score at a fixed 15%, with the seven components above carrying the other 85%. A resume can match a job perfectly on content and still lose points here for parsing badly. The ATS templates exist for exactly that problem.
Raising each component
- Keywords: read the posting for the terms it repeats, and add the ones that are genuinely true of you, in the role bullet where you did the thing. A skills list that is not corroborated by the experience below it is the weakest possible form of this.
- Title alignment: if your internal title is unusual, put the industry-standard equivalent alongside it. "Growth Ninja" matches nothing; "Growth Ninja (Marketing Manager)" matches.
- Recency: use one date format everywhere, write "Present" on the current role, and put the relevant experience where it will be read first.
- Achievements: take each bullet and ask what changed because you did it. If nothing changed, cut the bullet.
- Education and certifications: spell out the credential in full at least once, with the awarding body. A badge or an acronym alone is easy for a parser to miss.
Why this is published
A score you cannot interrogate is a score you cannot act on, and an opaque number that tells a qualified candidate they are a 43 is worse than no number. Everything above is the actual model: the same weights, the same cap, the same thresholds the analyser uses.
It will change as the model improves, and this page changes with it. If your score does not match what this page describes, that is a bug worth reporting rather than a nuance.
Related Articles
Keyword Optimization Guide
Learn how to optimize your resume with relevant keywords to pass ATS systems
Improving Your Resume Score
Actionable strategies to increase your ATS score and resume effectiveness
Comparing Resume Versions
Learn how to compare different versions of your resume and track improvements