Why your bilingual CV is failing the parser (and how to fix it in fifteen minutes)

You wrote a beautiful CV. Arabic on one side, English on the other, your name in an elegant header. Then you uploaded it to a job portal and the form auto-filled your job title as your phone number. You are not imagining it — bilingual documents break applicant tracking systems in predictable ways, and almost all of them are fixable in a single sitting.

We ran the same two-page bilingual CV through twelve parsers used by employers across Saudi Arabia and the UAE. Three failures showed up again and again.

Failure 1: The parser reads columns as one line

Most ATS engines flatten a page into a single stream of text before they look for fields. A two-column layout — Arabic on the right, English on the left — gets read straight across, so a line ends up as “Senior Product Manager مدير منتجات أول.” The parser then can't match your title to anything and drops it.

The fix is not to abandon two languages. It is to stop using real columns. Use a single-column layout and place each language in its own clearly-labelled section, one after the other. If you must keep two columns visually, keep the machine-read content in one column and treat the other as decoration the parser can ignore.

Failure 2: Right-to-left text corrupts numbers and dates

Phone numbers, dates, and salary figures placed inside a right-to-left paragraph often come out reversed or with digits regrouped. A number like +966 50 123 4567 can arrive as 4567 123 50 966+. Recruiters searching by phone or filtering by graduation year simply never find you.

• Keep contact details, dates, and any Latin-digit numbers in a left-to-right block, even on an Arabic CV.
• Write dates in an unambiguous format the parser can key on, e.g. 2024–2026, not ٢٠٢٤.
• Never let a phone number sit at the very start of an RTL line — anchor it with a label like Phone: first.

Failure 3: The font has no real Arabic glyphs

Decorative or Latin-first fonts frequently substitute Arabic letters from a fallback font, or worse, embed them as outlines with no underlying text. The CV looks perfect to your eyes and is completely invisible to the machine. Copy a paragraph of your exported PDF and paste it into a plain text field — if the Arabic comes back as boxes or nothing at all, the parser sees the same.

If you can't select and copy the text out of your own PDF, neither can the software deciding whether you get an interview.

The fifteen-minute fix

1. Switch to a single-column, ATS-safe template.
2. Move all contact details and dates into a left-to-right block with clear labels.
3. Use one font that ships genuine Arabic and Latin glyphs together.
4. Export to PDF, then copy-paste the whole thing into a plain text editor.
5. Read what came out. That garbled text is exactly what the employer's system stores — fix anything that isn't clean.

Do this once and you can reuse the same clean base for every application. The goal isn't a plainer CV — it's a CV that survives the trip through the machine with everything a human needs still intact.