Submission
You apply through a careers portal or a job board. The file and the form fields you filled in are both sent to the employer’s tracking system.
See what an applicant tracking system does with your file — formatting, structure, keyword coverage and measurable results — scored against the specific advertisement, not a generic template.
Interface preview. Nothing is uploaded or analysed yet — dropping a file shows its name only. The report below is a fixed sample, clearly marked, so you can review the layout end to end.
A plain explanation of what these tools measure, what they cannot see, and how much weight to give the number.
An ATS resume checker compares your resume against a specific job advertisement and reports how readable and how well matched it is. It does two separate jobs: a technical check on whether the file parses cleanly, and a language check on how much of the employer’s own vocabulary appears in your document.
Both matter, but for different reasons. A file that parses badly may never reach a recruiter’s search results at all. A file that parses perfectly but describes the same work in different words will reach the database and then never surface for the searches recruiters actually run.
The technical half is objective. Does the text extract in the right reading order? Are your name, email and phone found in the body rather than a page header? Do the section headings match labels parsers recognise? Are the dates in a consistent format that reads as a continuous history? These have right answers, and fixing them takes minutes.
The language half is comparative. The checker extracts terms from the advertisement — job title, tools, methods, qualifications — and reports which appear in your resume and which do not. It is a coverage report, not a verdict on whether you are qualified.
No checker can tell whether your experience is genuinely relevant, whether a career gap needs explaining, or whether your strongest achievement is buried on page two. It cannot judge tone, and it cannot tell that the phrase you matched describes something you did once, briefly, five years ago.
It also cannot see the parts of hiring that decide most outcomes: internal candidates, referrals, timing, and whether the role is genuinely open. A high score is a removed obstacle, not a prediction.
For any application you actually care about, yes — because the formatting problems it catches are invisible from your side. A resume can look immaculate on screen and arrive in a database with your job titles scrambled, and you would never know. That check costs a minute.
The keyword half is worth taking seriously and not literally. Match the advertisement’s language wherever it honestly describes your work. Where it does not, leave it alone — a term you cannot defend in an interview is worse than a missing one.
Six stages, in order. Only the last one involves a person reading your document.
You apply through a careers portal or a job board. The file and the form fields you filled in are both sent to the employer’s tracking system.
The system extracts text from the file and guesses which chunks are your name, employers, titles and dates. Reading order matters more than visual layout here.
Extracted values are written into structured database fields — one row per candidate, searchable by title, skill, location and dates.
A recruiter searches the database rather than reading every application. Your record either appears in their results or it does not.
Some systems order results by how closely a record matches the requisition. This is a sort, not a rejection — everyone stays in the database.
A person opens the shortlisted files and reads the first third of page one. This is where the actual decision is made.
Systems differ in detail, and most employers configure them differently. This is the shape common to nearly all of them.
Scores shown are from the example report above. Formatting and structure are within your control; keyword coverage depends on the specific advertisement.
Whether the file survives text extraction: reading order, columns, tables, embedded images and font encoding.
How much of the advertisement’s own vocabulary appears in your resume — the score most people are losing points on.
Whether the specific tools, methods and qualifications named as requirements appear anywhere in your document.
Sentence length, bullet density and how quickly a human can extract your value from the first third of page one.
Whether name, email, phone and location were found in the document body and parsed into the right fields.
Whether standard headings are present and recognised, and whether anything expected is missing.
Whether qualifications, institutions and dates parse into a structured record.
Whether roles, employers and date ranges form a continuous history, and how many bullets carry a measurable result.
Every one is invisible from the writer’s side, and every one takes minutes to fix.
Six steps in the order that saves the most time. The first two do most of the work.
Run the checker once with no changes. Formatting and contact-information scores are objective and usually take under ten minutes to bring above 90. Do these first — there is no point tuning language in a file that parses badly.
If the advertisement says “Product Manager” and your resume says “Product Lead”, add the recognised title. This single edit moves keyword coverage more than any other.
Copy the “what you’ll need” section into a list. Mark each item as: I have this and it is on my resume, I have this and it is missing, or I do not have this. Only the middle group needs work.
Add them inside an achievement, not to a skills wall. “Ran quarterly forecasting in Excel and Looker” is credible; a comma in a list is not.
Any bullet without scale, timeframe or outcome is a duty. Aim for a figure in most of them — team size and volume count.
Above 85 on formatting, structure and contact information every time. Keyword coverage varies by how well you fit the role, and chasing 100 means overstating something.
A score is not a prediction. It tells you your resume is likely to be read, not that it will win. Screening decisions are made by people weighing experience, timing and internal candidates — see the disclaimer.
Generic checkers grade you against a template. This one compares your file with the specific job ad, which is the only comparison that changes what you should edit.
Checks run in your browser. Your resume, your employment history and the job you are applying for never leave the page.
Every check says what is wrong and what to do about it. No score without a reason, and no jargon you have to look up.
Fixes you make flow straight into the resume builder, which already produces the layout the checker rewards.
No photo, no date of birth, reverse chronological order — the conventions those markets share, with warnings where local practice differs.
No sign-up, no credit card and no email wall in front of the result. Report export is the only thing that needs an account.
Six situations, each with a different first move.
The situation: Applying to graduate schemes that run everything through a portal.
Where to start: Check formatting first — with a short history, a parsing error removes most of what you have.
The situation: Competing against everyone in your year with a similar degree.
Where to start: Coverage is where you win. Match the tools and methods named in the ad, using your project work as evidence.
The situation: Sending the same file to every opening and wondering why it goes quiet.
Where to start: Run each application separately. The score difference between two ads for the same title is often twenty points.
The situation: Senior roles are often filled by search, but portals still gate the applications.
Where to start: Check that scope and outcomes parse. Two pages, impact first, everything older than a decade on one line.
The situation: Twelve clients in three years confuses a system expecting one employer per date range.
Where to start: Group under a single practice entry so the history parses as continuous rather than fragmented.
The situation: Your experience is relevant; the vocabulary belongs to a different industry.
Where to start: Coverage will start low, and that is the point. Use the missing-terms list as a translation checklist.
Whether the file survives extraction: reading order, fonts, columns, tables and images.
Whether standard headings exist and dates parse into a continuous history.
How much of the ad's own language appears in your resume — the number most people are missing.
The share of bullets carrying a number, a scale or a timeframe rather than a duty.
No. Hidden keyword stuffing is detected by most modern systems and reads as dishonest to the human who sees the file afterwards. Match the ad's language only where it describes work you've actually done.
Above 85 on formatting, structure and contact information every time, because those are entirely within your control. Keyword coverage varies with how closely your background fits the role, and pushing it to 100 usually means claiming something you cannot defend.
Yes. Checking is free with no account and no card. Downloading the report as a file is the only part that opens with an account.
No. In this build nothing leaves your browser — dropping a file shows its name only. When analysis ships it will run locally, and anything stored will be opt-in and covered by the privacy policy.
PDF, DOCX and plain text. Export a text-based PDF rather than scanning a printed page — a scan contains no extractable text at all.
No. It means your resume is likely to be read rather than lost in a database. Screening decisions are made by people weighing experience, timing and internal candidates.
Rarely, despite what most advice implies. In common configurations they sort and filter rather than reject. The realistic risk is a parsing error that keeps you out of a recruiter’s search results.
Yes, where they honestly describe work you have done — recruiters search the database using the advertisement’s own words. Never add a term you could not discuss in an interview.
No. Hidden white text and skill walls are detected by modern systems, and they read as dishonest to the person who opens the file afterwards. The downside is far larger than the upside.
It is a risk you cannot measure in advance. Some parsers read straight across the page and scramble the order. Use a single column when applying through a portal; keep the designed version for a human.
Not for the US, UK, Canada or Australia, where it raises discrimination concerns and is often stripped. Parts of Europe still expect one — check the convention where you are applying.
Once per application you care about. Two advertisements for the same job title routinely produce scores twenty points apart, because they use different words.
Yes. The checks reflect conventions shared across the US, UK, Canada, Australia and much of Europe: reverse chronological order, no photo, no date of birth, standard section headings.
Above 85 on formatting and structure every time — those are within your control. Keyword coverage varies by how closely your background fits, and chasing 100 usually means bending the truth.
It means your resume is likely to be read, not that it will win. Screening decisions are made by people weighing experience, timing and internal candidates — see the disclaimer.
One minute of checking removes the formatting problems you cannot see from your side.