ATS guides
Appliora CV Readability & ATS Parsing Study 2026
A reproducible snapshot of how 20 Appliora CV layouts behave when the same document is exported and read by the parser shipped in the free ATS Checker. The study is about text recovery and reading order, not hiring outcomes or competitor products.
Key findings
The result is useful because it is allowed to be unflattering.
- 14 of 20 tested layouts triggered the parser's column reading-order finding.
- 6 of 20 layouts had no critical parser finding in this run.
- The observed parse rates range from 60% to 100%; the score can be capped when reading order is scrambled.
Methodology
The run is designed so another engineer can reproduce it.
- One fixed fictional CV was loaded into every tested layout; content was not changed between rows.
- Each layout was printed to PDF by headless Chrome with a real text layer.
- The running site's shipped extraction and scoring modules were used; no separate reimplementation was used.
- The scoring model was version 4.5; the run date is 2026-09-18.
- Run command: npm run seo:benchmark. It rewrites content/data/parsing-benchmark.json, which is the source for this page.
Tested document and template set
The document is the shared ATS-focus persona used by the benchmark harness. It contains contact details, summary, work history, education, skills and quantified examples. The tested set is 20 layouts from the active collection; the raw JSON records every label, internal id, score, parse rate and finding.
Individual checks
The parser scores text-layer integrity, contact extraction, section recognition, dates, bullets, skills, file hygiene, spelling, diacritics, inclusion risk, length and language quality. The reading-order finding (parse.columns) is reported separately because a document can contain all its words and still present them in the wrong sequence.
Results
| Template | Score | Parse rate | Reading order | Other findings |
|---|---|---|---|---|
| Dark Modern | 97 | 100% | clean | 2 |
| ATS Structured | 95 | 100% | clean | 3 |
| Product Systems | 95 | 100% | clean | 3 |
| Academic / Research | 94 | 100% | clean | 3 |
| ATS Clean | 92 | 100% | clean | 4 |
| Student / First Job | 90 | 100% | clean | 4 |
| Bold Modern | 69 | 71% | scrambled | 8 |
| Commercial | 69 | 83% | scrambled | 3 |
| Creative Portfolio | 69 | 89% | scrambled | 5 |
| Dark Sidebar | 69 | 79% | scrambled | 4 |
| European | 69 | 60% | scrambled | 4 |
| Executive Premium | 69 | 76% | scrambled | 3 |
| Grand Hotel | 69 | 68% | scrambled | 4 |
| Growth Marketing | 69 | 82% | scrambled | 2 |
| Hospitality Service | 69 | 79% | scrambled | 5 |
| Light Sidebar | 69 | 78% | scrambled | 2 |
| Medical / Healthcare | 69 | 77% | scrambled | 4 |
| Swiss One Page | 69 | 67% | scrambled | 4 |
| Tech Stack | 69 | 88% | scrambled | 4 |
| Two Column Pro | 69 | 82% | scrambled | 4 |
Source: content/data/parsing-benchmark.json. Run 2026-09-18; model 4.5.
Factual conclusions
In this snapshot, multi-column layouts were the main source of reading-order failures, while clean parsing did not guarantee a high content score. The study supports checking the exported text layer and reading order before treating a visual CV as ATS-ready. It does not support claims about any employer's ATS or interview probability.
Limitations
- One fictional CV cannot represent every career or document length.
- The benchmark measures Appliora's parser, not Workday, Greenhouse, SAP SuccessFactors or another closed ATS.
- Headless Chrome print-to-PDF is not identical to every builder export path.
- Templates and parser rules change; this is a dated snapshot, not a permanent certification.
Raw data and reuse
Download or inspect the raw dataset in the repository: content/data/parsing-benchmark.json. You may cite this page as the Appliora CV Readability & ATS Parsing Study 2026 with the run date and model version. Please link the methodology and limitations with the citation.
FAQ
Does a clean parse mean a CV will pass an ATS?
No. It means this parser recovered the text in order and found no critical issue in this run.
Did you test competitors?
No. Only Appliora layouts are included, so no competitor claim is implied.
See what a parser reads from your CV
Upload your current file and get the extracted text plus a 0-100 score across 14 checks. No signup.