The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →In a test reported by ResuHive, one of six open-source resume-template PDFs produced no Arabic numerals when processed by the publisher’s resume-upload text extractor. The Deedy Chinese edition yielded 1,709 extracted characters, but none of the dates, metrics, or other digits visible in the rendered PDF appeared in that text. The tested English Deedy two-column PDF, by contrast, passed the publisher’s layout checks. These are results from one parser workflow—not a verdict on every template, PDF export, or applicant tracking system (ATS).
What the six-template test found
ResuHive says it ran six PDFs through the same extraction library and call used by its resume uploader, then checked the extracted text with the rules behind its free ATS check. The parsing run was dated September 14, 2026. The publisher names its audit script as scripts/ats-template-audit.ts and says it recorded each template’s source, license, and commit alongside the files. The available report does not expose enough detail to independently verify the exact files, commits, parser version, or script.
As an Amazon Associate I earn from qualifying purchases.
| Template listed in the report | Layout reported | License reported | Digits recovered in that run |
|---|---|---|---|
| sb2nov | Single column | MIT | 76 |
| Awesome-CV (resume) | Single column | LPPL-1.3c | 255 |
| Awesome-CV (cv) | Single column | LPPL-1.3c | 453 |
| Deedy | Two columns | Apache-2.0 | 158 |
| zheyuye (Chinese) | Single column | MIT | 69 |
| Deedy Chinese edition | Two columns | Apache-2.0 | 0 |
The template names, license labels, layouts, and counts are as reported by ResuHive; they have not been independently reproduced. The counts describe this test run, not a general ranking or a prediction of how another parser will behave.
Free tools Windows power users keep installed
One-click scans. No signup required.
Why the Deedy Chinese PDF lost its digits
ResuHive reports that the Deedy Chinese edition’s PDF text layer contained 1,709 characters but zero Arabic numerals. Numbers remained visible on the rendered page, while the extracted text omitted them. That matters because dates and quantified accomplishments can be essential information in a resume, even when a human reader can see them on screen.
#1 Best Overall
The same extraction reportedly substituted 69 visually similar Kangxi radicals for characters. For example, it returned ⾼ (U+2F98) instead of 高 (U+9AD8); the report also describes a radical-form substitution in 工程. Because those are distinct code points, an exact-match search for the standard character may not find the visually similar one. This is a reported result for this PDF and extraction workflow, not proof that all copies or exports of the template behave this way.
Does the result mean two-column resumes fail ATS checks?
No. In the same report, the English Deedy two-column PDF passed ResuHive’s layout checks: its email and phone were extracted, contact details appeared near the start, and no lines were flagged for scrambled column bleed. That is evidence only about the tested PDF and workflow. It does not establish that every ATS will read every two-column design correctly, nor that tables, text boxes, or other layouts will never interleave during extraction.
The contrast is a useful reminder: visible column count alone does not establish whether extracted text is usable. The test also shows that a file can retain readable-looking page content while its programmatically extracted text has serious character losses.
Recommended Free Tools
How to check whether your resume PDF preserves dates and numbers
- Open the PDF and copy its text. Select the resume text, copy it, and paste it into a plain-text destination such as a basic text editor.
- Inspect dates and metrics. Check job dates, education dates, phone numbers, percentages, quantities, and other numerals against the rendered PDF. Do not assume that text visible on the page is present in the extracted version.
- Check non-Latin text and symbols. Look for missing characters or visually similar substitutions, especially if correct spelling or exact-match search matters.
- Check reading order. Confirm that contact details and sections appear in a sensible sequence and that text from columns has not become interleaved.
- If anything is missing, try another export. ResuHive suggests re-exporting with fonts fully embedded or using a different export path. The report does not independently establish that either change fixes every PDF, so repeat the copy-and-paste check on the new file.
For a more meaningful comparison between versions, keep the resume content and parser workflow constant where possible. Compare the extracted dates and digits, exact characters, contact and section order, and any column interleaving—not just how the PDF looks.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What this test can—and cannot—tell job seekers
The strongest conclusion is narrow: one tested Deedy Chinese edition produced extracted text without digits and with reported Unicode substitutions, while the tested English Deedy two-column PDF passed the publisher’s layout checks. The five other files returned between 69 and 453 digits in that run. A digit count by itself does not prove that every number was preserved or that the resulting text was otherwise accurate.
- It can show: how these six PDFs behaved in the publisher’s reported extraction workflow, including one text-layer failure and one two-column file that passed its own checks.
- It cannot show: how all ATS products handle the same files, how other exports or versions behave, or whether any result affects hiring outcomes.
- It does not establish: that all two-column resumes are unsafe, that all single-column resumes are reliable, or that the suggested export changes will always resolve missing text.
ResuHive’s report is the source for the findings and describes a sample, high-level method, and excluded false alarm. It does not provide independent replication or a second parser’s results. The zheyuye sample used 13xxx-xxx-xxx as a deliberate phone-number placeholder, which the publisher excluded as a false missing-phone flag; sample content can therefore affect automated checks.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →
