Included in every plan · Free extraction preview
ATS Resume Checker
An Applicant Tracking System has to extract your resume before a recruiter can search or rank it. Upload yours, get a parsing score, and see every block your current draft is hitting.
What we actually check
Parseable structure
Column layouts, text-in-image, exotic fonts, and embedded tables that strip your content before a human sees it.
Section hygiene
Standard headings the ATS recognises (Experience, Education, Skills) vs custom ones it skips.
Keyword coverage
How well your resume matches the job description's hard skills and tooling — not just generic buzzwords.
Date + role format
Consistent date ranges, employer-then-title order, and role-level signals the parser relies on.
Bullet quality
Verbs, metrics, and length — flags walls of text and bullets that say nothing measurable.
Filename + metadata
PDF metadata, filename hygiene, and file format issues that some ATS reject before parsing.
What is an ATS resume checker?
An ATS resume checker reads your CV the way an applicant tracking system does — as a parsing job, not as a document — and reports what survives that process. It converts the file to plain text, tries to split it into structured fields (name, contact, employment history, education, skills), and shows you which parts came out empty, garbled or in the wrong field.
That is a different question from “is my CV good?”. A CV can be well written, well designed, and still arrive at the recruiter as three columns of interleaved nonsense. The checker answers only the first question: does the machine in the middle read this correctly?
How an ATS actually reads your CV
Applicant tracking systems differ in detail, but the pipeline is broadly the same everywhere, and knowing the order explains most failures.
1. Text extraction
The file is converted to raw text. A PDF built from vector text extracts cleanly. A PDF that is really a scanned image extracts nothing at all without OCR — the system sees a blank document. This is the single most destructive failure, and it is invisible to you because the file looks perfect on screen.
2. Reading order
Text is read in the order it exists in the file, which is not always the order you see. Two-column layouts are the classic case: the parser may read straight across both columns, interleaving your skills list into the middle of a job description.
3. Section segmentation
The parser looks for headings it recognises to decide where employment history starts and ends. “Experience”, “Work Experience” and “Employment” are understood almost everywhere. “Where I've Made an Impact” is not — and everything under it may be dropped or filed as uncategorised text.
4. Field mapping
Inside each section it extracts entities: employer, job title, start date, end date. This is where inconsistent date formats break things. Mixing “Jan 2023 – Present” with “03/2021-12/2022” in the same CV can produce roles with missing durations.
5. Keyword and skill matching
Only then is the extracted text compared against the job requisition. Note the order: if step 1 or 3 failed, your skills were never in the text being matched. You do not lose on relevance — you lose before relevance is considered.
The formatting choices that break extraction
Almost every serious parsing failure traces back to a small set of design decisions. In rough order of how much damage they do:
| Choice | What the parser gets | Do instead |
|---|---|---|
| CV exported as an image, or a scanned PDF | Nothing. An empty document. | Export text-based PDF or DOCX from the original file. |
| Two- or three-column layout | Columns interleaved line by line, sentences spliced together. | Single column, top to bottom. |
| Contact details in the header/footer | Often skipped entirely — some parsers never read header regions. | Put name, email and phone in the body, at the top. |
| Skills inside a table or text box | Table cells flattened unpredictably; text boxes sometimes dropped. | A plain bulleted or comma-separated list. |
| Custom section headings | Section unrecognised; contents filed as uncategorised. | Experience, Education, Skills — the boring words. |
| Icons or symbols standing in for labels | A phone glyph carries no text; the number loses its meaning. | Write the word: “Phone:”, “Email:”. |
| Inconsistent date formats | Roles with missing or wrong durations. | One format throughout, e.g. “Mar 2022 – Jun 2024”. |
| Unusual or embedded fonts | Character substitution, ligature errors, mojibake. | A standard system font. |
How to read your score
A single number is a summary, not a verdict, and the components matter far more than the total. Two CVs can both score 70 for opposite reasons — one parses perfectly but misses the role’s vocabulary, the other is keyword-rich but half-unreadable. They need completely different fixes.
- Parsing failures first. If a section came out empty, fix that before touching anything else. Keywords cannot help text that was never extracted.
- Then section coverage. Every standard section present and correctly identified.
- Then keyword alignment against the specific posting — not against the job title in general. This is the part that has to be redone per application.
- Ignore cosmetic points. Length and bullet-count heuristics are the weakest signals any scorer produces. Do not restructure a good CV to chase them.
Common questions
- Is a PDF or a Word document safer?
- Both parse well when the file contains real text. DOCX is marginally safer as a default because it cannot accidentally be an image, and because a few older systems handle it more predictably. A text-based PDF exported from the same source is fine almost everywhere. The format matters far less than whether the text is extractable and the layout is single-column.
- Does an ATS reject my CV automatically?
- Usually not on its own. In most configurations the system ranks and filters, and a human reviews what surfaces. The practical risk is not a hard rejection but never appearing high enough in the list to be looked at — which is the same outcome from your side, and is exactly what parsing failures cause.
- Should I put keywords in white text to game the filter?
- No. Hidden text is trivially detected, it appears in the extracted plain text a recruiter can view, and it reads as deception rather than cleverness. It is one of the few things that will genuinely get an application discarded.
- How often should I re-check?
- Re-check the parsing once, after any change to layout, template or export method. Re-check keyword alignment for every application, because that part is relative to the posting and nothing else.
- Do I need a different CV for every job?
- You need the same underlying CV with the vocabulary matched to each posting. The structure, roles and achievements stay put; the phrasing of skills and the emphasis shift toward what the requisition actually asks for.
Why use Lumyhired’s checker
Other ATS scorers give you a vague number. We surface the specific line, the specific section, and the fix — then offer to apply it automatically. The checker is part of the weekly plans, from €5 with a 3-day money-back guarantee.
If you want the reasoning behind all of this in one place, the full guide to passing ATS filters goes deeper. You can also see how an ATS reads your resume before you upload it, and the resume examples show the target shape for your role.
Want the raw extraction first? The ATS extraction preview runs in your browser, with no account and nothing uploaded.