A resume ATS checker parses your file the way an applicant tracking system would, then scores two things: whether the structure is readable and how well your language overlaps with a target role. It's the closest free way to see your resume the way software — not a person — sees it first.
This guide breaks down exactly what happens when you run one, which specific formatting choices break parsing on real platforms like Workday and Taleo, and how to use the results without over-trusting a single number.
How a resume ATS checker actually works
Key Takeaway
It parses, then maps, then compares — the same three-step process real ATS platforms use before a human ever opens your file.
- Parse. The checker extracts raw text from your PDF or DOCX, the same way Workday or Greenhouse would on upload. This step converts your visually formatted document into a flat text string, stripping away fonts, colors, and layout.
- Map. It tries to identify structured fields: name, contact, job titles, dates, employers, education, skills. This is done with pattern matching and positional logic — the parser looks for a phone-number-shaped string, a date-range-shaped string, and section headers like "Experience" or "Education."
- Compare. If you provide a job description, it measures keyword and skill overlap between your resume and that specific posting, using both exact-text matches and near-synonym matching (e.g., "managed" vs. "led").
Everything that follows — your score, your flags, your suggestions — comes from how cleanly those three steps completed. If step one fails (the parser can't extract clean text), steps two and three inherit that failure: a garbled parse produces garbled field mapping, which produces a meaningless keyword comparison. And if you're wondering whether you need to sign up for any of this, you don't .
What goes wrong if you skip understanding this: people assume a "bad score" means their career history is weak, when in nine cases out of ten it means step one or step two failed on formatting — a fixable five-minute problem, not a rewrite-your-whole-resume problem.
How different ATS platforms parse resumes differently
Key Takeaway
Not all ATS platforms parse the same way — older systems like Taleo are far less forgiving of formatting than modern ones like Lever or Greenhouse.
"The ATS" isn't one piece of software — it's a category of dozens of products with very different parsing engines. Consumer checkers approximate common behavior across the most widely deployed platforms, but knowing how the big ones differ helps you understand why formatting advice sometimes feels overly cautious.
| Platform | Known parsing behavior |
|---|---|
| Taleo (Oracle) | A legacy enterprise system still used by many large companies. Its parser is notoriously literal — multi-column layouts, tables, and text boxes frequently scramble field order or drop content entirely. |
| Workday | Widely used by mid-to-large employers. Generally reads single-column resumes well but has a well-documented history of misreading dates in non-standard formats and struggling with resumes that use graphics for skill ratings. |
| Greenhouse | Common at startups and tech companies. Uses a more modern parsing layer and typically extracts standard sections reliably, though it still fails on image-based PDFs and heavy tables. |
| Lever | Also popular with tech and growth-stage companies. Parses cleanly for most single-column resumes and tends to map contact fields accurately as long as they're in the body text, not a header/footer. |
| iCIMS | Common in retail, healthcare, and staffing. Reliable for plain-text-heavy resumes; like Taleo, it can struggle with decorative elements like icons next to contact details. |
You won't know in advance which platform a specific employer runs. That's exactly why the safest strategy is to format for the least forgiving system (Taleo-style plain structure) rather than betting on the most forgiving one.
What it catches vs. what it misses
Key Takeaway
It's excellent at structural and keyword problems. It can't judge your actual qualifications or writing quality.
| Catches reliably | Can't evaluate |
|---|---|
| Multi-column layouts that scramble parsing | Whether your experience is actually strong |
| Contact info hidden in a header/footer | Career narrative coherence |
| Missing keyword overlap with a job description | How a specific hiring manager will react |
| Non-selectable (image-only) PDF text | Competition level for the role |
| Tables and text boxes that drop content on parse | Whether your bullets are believable or exaggerated |
What goes wrong if you skip this distinction: people either dismiss the tool entirely ("it can't judge my career, so it's useless") or over-trust it ("it said 92, I'm definitely getting an interview"). Both reactions miss that it's a narrow, specific instrument — good at exactly two things, and honest about not doing a third.
Resume elements that break ATS parsing
Key Takeaway
Most parsing failures trace back to one of six specific layout choices — all of them fixable without losing visual quality.
- Multi-column layouts. Parsers read left-to-right, top-to-bottom in a single pass. A two-column resume gets its columns interleaved line by line, so a job title from the left column can end up glued to a date from the right column.
- Text inside tables. Table cells are sometimes skipped entirely or extracted in the wrong reading order, especially in Taleo and iCIMS.
- Header and footer placement. Contact info placed in a document header or footer (rather than the body) is frequently ignored by parsers that only scan the main content flow — this is the single most common cause of a "missing phone number" flag.
- Graphics for skills or ratings. Progress bars, star ratings, and skill icons carry no extractable text — a parser sees an empty box where "Python: Advanced" should be.
- Non-standard date formats. "Jan '22 – Present" or "1/22 - Current" can fail to match the date-range pattern a parser expects, causing employment gaps to be miscalculated or fields to map incorrectly.
- Scanned or image-based PDFs. If you can't select and copy the text in your own PDF viewer, no ATS on earth can read it either — there's no text layer to extract.
If you skip fixing these: your resume can look flawless on screen and still parse into a scrambled, incomplete record inside the employer's actual system — invisible to you, and often invisible to the recruiter too, since they're seeing the parsed profile, not your PDF.
Before and after: three real fixes
Key Takeaway
The biggest score jumps almost always come from structure, not from adding more content.
Before (score: 44): Two-column template, phone number in a sidebar text box, skills shown as colored progress bars.
After (score: 87): Single-column layout, phone number moved into the body under the name, skills rewritten as a plain text list.
Before: "Responsible for team reporting and communications."
After: "Built weekly reporting for a 12-person team, cutting status-update time from 3 hours to 40 minutes using a shared dashboard."
Before (dates: "Summer 2021 – Present"): Parser couldn't match "Summer 2021" to a month, leaving the role's duration blank in the mapped record.
After (dates: "Jun 2021 – Present"): Standard month-year format parsed cleanly, and total tenure calculated correctly for the first time.
None of these three people changed their actual career — they changed how legibly it was presented, both to the software and to the person reading it six seconds later.
See your own structure and keyword score now.
Run HireFlow's free resume ATS checker — upload a PDF or DOCX, no signup, results in about 30 seconds.
How to use a resume ATS checker step by step
Key Takeaway
Fix structure first, keywords second, and stop once you're honestly covered — don't chase a perfect score.
- Upload your resume as a PDF or DOCX and get a baseline structure score. Skip the JPEG/PNG export option some design tools offer by default — image exports have no text layer.
- Fix every structural flag first. These are almost always the biggest score movers, often worth 20-40 points on their own, because they unblock accurate field mapping for every step after.
- Paste the specific job description you're targeting to see real keyword gaps for that role. A generic "software engineer" scan tells you far less than pasting the actual posting from the company you're applying to.
- Add honest overlap where it's genuinely true. If the posting says "stakeholder management" and you've done that under a different label, rename it — don't paste posting language verbatim for skills you haven't actually used.
- Re-scan once. In the mid-70s to high-80s with no critical flags, apply. Diminishing returns set in fast after that point.
How keyword matching actually works
Key Takeaway
Matching isn't just exact-string search — but it's also not smart enough to infer skills you never wrote down.
Modern parsers use a mix of exact-text matching and a synonym or "skills taxonomy" layer that maps related terms together — recognizing that "Excel" and "Microsoft Excel," or "led" and "managed," refer to the same thing. This is more forgiving than older keyword-stuffing advice assumes.
What it won't do is infer a skill from context. If a job description asks for "SQL" and your resume describes building "data pipelines" without ever naming SQL, a parser generally won't connect those dots the way a human reader might. Name the specific tools and skills literally, at least once, even if you also describe them narratively elsewhere.
What goes wrong if you skip this: candidates with genuinely relevant experience get filtered out for keyword coverage, not lack of ability — the exact scenario a resume ATS checker exists to prevent.
Mistakes people make with the results
Key Takeaway
A number without action is useless — and an over-optimized number can actively hurt you with a human reader.
- Chasing 100 through keyword stuffing. A recruiter notices a Skills section longer than your Experience.
- Running one generic check and applying everywhere. Keyword overlap changes per job description — structure doesn't need to be re-checked, but relevance does.
- Ignoring structural flags because the score still "passed." A layout flag can still silently drop a section in a different employer's system.
- Treating the score as the finish line. It's a rehearsal for parsing, not a verdict on your candidacy.
- Assuming one platform's forgiving parser means all of them are. A resume that parses fine on Lever can still fail on Taleo — format for the strictest case, not the most lenient.
Frequently asked questions
It parses your resume file the way a real applicant tracking system would, then reports on two things: whether the structure is readable (contact info, dates, job titles landing in the right fields) and how well your language overlaps with a target role. It's a diagnostic, not a hiring decision.
No. Consumer checkers approximate common parsing and matching behavior seen in platforms like Workday, Greenhouse, and Taleo. They don't have access to any specific employer's live system, so results are a strong directional signal, not a guarantee of what one company's exact software will do.
Yes, as long as the PDF has a real text layer — meaning you can select and copy text from it. A scanned image saved as a PDF has no text layer, and neither a checker nor a real ATS can read it. Export directly from Word or Google Docs rather than scanning a printed copy.
Roughly 75-90 with no critical structure flags is a realistic, healthy target. Below 60 usually points to a layout problem. Be skeptical of chasing a perfect 100 — it can sometimes come from keyword stuffing rather than genuine quality.
No. It confirms your file is likely to parse correctly and shows reasonable keyword overlap with a role. It can't account for how many other people applied, your actual qualifications, or a specific hiring manager's preferences.
Run a full structural check once when you finalize your resume template. After that, a quick keyword check against each new job description is worth the 30 seconds it takes — structure rarely changes between applications, but keyword overlap does.
Yes, and arguably they matter more here. Career changers often use non-standard job titles or industry jargon that doesn't map cleanly to a new field's expected keywords — a checker will surface that gap directly instead of you finding out through silence.
Structure scores stay stable between checks since your file layout hasn't changed. Keyword-match scores move with every job description because they measure overlap with that specific posting — a lower score against a new posting means the required skills genuinely differ, not that something broke.
Done for you
Turn this advice into an interview-ready resume
Professional writers rebuild your resume for ATS + recruiters — unlimited revisions, interview guarantee.