Free ATS Resume Checker — what ATS systems actually look for
Most resumes fail before a human reads them — the ATS strips them, mis-parses them, or scores them too low to surface. Here's what applicant tracking systems actually do, the common killers, and what AdaptMyCV checks during the Adapt flow.
Free with your account — no coins charged — try the free checker now → Or skip ahead to the full AI-tailored Adapt flow (10 coins ≈ $2, with 10 free coins on signup).
How ATS systems actually work
An ATS is software that sits between you and the recruiter. Workday, Greenhouse, Lever, iCIMS, Taleo, BambooHR, SmartRecruiters — different vendors, similar shape:
- Ingest. You upload a PDF or DOCX. The ATS extracts text using a parser. PDFs that look identical on screen can parse very differently — a single-column layout is clean, a two-column layout often interleaves text from both columns, an image-only PDF extracts nothing.
- Field extraction. The parser tags chunks as name, contact, education, experience, skills. It looks for standard section headers — “Experience”, “Work history”, “Education” — and date patterns to anchor work entries.
- Scoring. Some ATSes do keyword matching against the job description; others compute a more elaborate score including years-of-experience inference, education level, and location. Either way, low overlap with the JD's keywords hurts.
- Ranking. Recruiters typically see the top N candidates by score. If you're #82 of 300, you're almost certainly not getting a human review.
Common ATS killers
- Multi-column layouts. The parser reads in reading order across the whole page, so “senior engineer” from the left column gets concatenated with “awards: hackathon” from the right.
- Text inside images. Logos and decorative banners are fine if redundant with the text. An image where the job title or company name lives only as pixels is invisible to the parser.
- Unusual fonts. Most parsers are forgiving, but non-Latin fonts and ornamental typefaces sometimes return garbage. Stick to system fonts: Inter, Helvetica, Arial, Calibri, Times New Roman, Garamond.
- Tables for layout. Borderless tables often parse cleanly; tables-of-tables nested for visual effect rarely do.
- Headers and footers with key info. Some parsers ignore the header/footer regions entirely. Don't put your contact details only in the header.
- Non-standard section names. “My journey” instead of “Experience” might not get tagged as experience. Default to predictable section headers.
- Unicode dates. “Jan 2020 — Present” with an em-dash usually parses; rare unicode separators sometimes don't.
- Keyword absence. If the JD asks for “CI/CD” six times and your CV says “build pipelines” once, your score is low. The fix is to mirror the JD's actual phrasing — honestly — where it applies.
What AdaptMyCV checks during Adapt
When you adapt your CV to a job description, AdaptMyCV runs an analyser before the rewrite:
- Keyword extraction from the JD — hard skills, tools, certifications, methodologies, seniority signals. Classified into “must-have” vs “nice-to-have.”
- Match scoring against your current CV: which keywords are already present, which are defensibly inferable, which are genuinely missing.
- Coverage report after the rewrite — the percentage of priority keywords that landed on the page.
- Honesty check on every line of the rewrite to make sure nothing was fabricated. (see how.)
- Anti-stuffing pass that strips marketing-style keyword tails (“..., utilizing X”, “..., demonstrating Y”) — those signal stuffing to recruiters even when the ATS scores them positively.
Check by role
ATS keyword expectations differ by role. Pick yours to see the specific keywords recruiters expect and the common mistakes for that job family.