What the score measures
The checker returns one overall score out of 100 and four category scores: keywords, formatting, content, and structure. Those four are the only dimensions scored. There is no hidden fifth factor and no employer-specific model behind it.
How to check this: run the same resume through the ATS checker and count the categories returned. If you see anything other than those four, this page is out of date and we want to know.
Structure and extraction
Templates use a single column, familiar section names, selectable text, and predictable reading order. We test exported PDFs for text extraction, multi-page wrapping, empty fields, long content, and contact links.
How to check this: export a PDF from the builder, then open it in any text extractor — including copy-and-paste into a plain text editor. The text should come out in the order you read it, with headings intact and nothing missing. If it does not, the template has a defect and the claim above is false.
What the score cannot tell you
It cannot predict an interview or a hire, and we will not claim otherwise. There is no single applicant tracking system: Workday, Greenhouse, Taleo, iCIMS and Lever behave differently, and each employer layers its own screening rules and human judgment on top. We have no access to any employer’s configuration, so no score here can be calibrated against the system you are actually applying into.
We also do not repeat the widely circulated claim that a fixed percentage of resumes are rejected before a human sees them. We cannot trace that figure to a credible study and applicant tracking vendors dispute it. Where we cannot source a number, we leave it out.
The analysis is generated, and it varies
Scores and suggestions are produced with AI assistance, not a fixed rule engine. Two consequences follow, and both are worth knowing before you act on a result. Scoring the same resume again after a material edit can move the score by a few points for reasons that have nothing to do with your edit. And an identical resume and job description submitted within an hour returns a cached result, so an unchanged re-run will give you exactly the same answer rather than a second opinion.
How to check this: submit the same resume twice in a row — the second result will be identical. Change one word and submit again to see fresh analysis. Treat the score as a diagnostic that points at weak sections, not as a measurement with two-decimal precision.
Privacy
Resume content and contact details are never sent to analytics. Analytics events carry only a page path, a page type, and scalar values such as whether a job description was supplied; they do not fire at all unless you choose “Allow analytics”, and IP anonymization is enabled. Resume text is sent to our AI provider to produce the analysis, which is the only way the feature can work, and it is not sold or used for advertising. Email delivery is transactional and does not subscribe you to marketing.
How to check this: open your browser’s network tab, decline analytics, and use the builder. You should see no requests to Google Analytics at all. Accept analytics and the events you see should contain no resume text.
Corrections
If any claim on this page is wrong, it is a defect and we would rather hear about it than keep it. Our editorial policy covers how corrections are handled, and the contact page reaches us.