Scanned CVs — photographed or scanned documents sent as image files — are one of the hardest inbound formats for recruitment agencies to process. Here's how to reformat them into clean, branded, client-ready submissions.
A candidate sends their CV as a photo. Or as a scanned PDF. Or as a JPEG taken with their phone.
You can't copy text from it. You can't edit it in Word. Your usual formatting workflow doesn't work — because the "CV" is technically a picture of a document, not a document at all.
Scanned CVs are common in specific markets — parts of Africa, the Middle East, Southeast Asia, and among older candidates in every geography. For recruitment agencies operating internationally, they're a regular occurrence. This guide covers what makes them hard to reformat, and how to convert scanned or image-based CVs into clean, branded client submissions reliably.
Why Scanned CVs Are a Different Problem
A "regular" CV — even a badly formatted one — contains extractable text. You can copy-paste it. Search within it. Highlight sections. The document contains actual text data that formatting tools can parse.
A scanned CV doesn't contain any of that. It's a picture. The words visible on the page are made of pixels, not characters. To your computer, it's the same category of file as a photograph of a sunset — it has no idea there's text in the image.
To turn a scanned CV into a reformattable document, you need OCR — optical character recognition. OCR is the technology that identifies text inside images and extracts it into machine-readable data.
OCR technology has improved significantly in the last few years, but not all formatting tools have kept up. Many CV formatting platforms simply reject image-based inputs — telling recruiters "please ask the candidate to send a Word file" — which introduces delay and can feel unprofessional.
When Recruiters Encounter Scanned CVs
Scanned or image-based CVs typically arrive in a few scenarios:
Photos from mobile devices. A candidate takes a picture of their printed CV and sends it via WhatsApp, email attachment, or LinkedIn message. Very common in markets where physical CVs are still the norm.
PDF scans of printed CVs. A candidate scans a printed copy of their CV — often at a print shop or with a home scanner — and sends the scan as a PDF. These look like PDFs but contain no extractable text.
Screenshots. A candidate takes a screenshot of their LinkedIn profile or another CV and sends the image directly.
Old or damaged documents. A candidate submits a photograph of an older printed CV where re-typing it in Word would be more effort than they want to invest.
All of these are essentially the same problem from a formatting perspective: image content, no text data, requires OCR to process.
How to Format a Scanned CV
The reliable workflow for handling a scanned CV is:
Step 1: Upload the image or scanned PDF to a CV formatting tool that supports OCR.
CVFormatter accepts image files (PNG, JPEG, JPG) and scanned PDFs directly. No preprocessing needed. No asking the candidate to re-type or resend in a different format.
Step 2: OCR extracts the text.
The system reads the image, identifies text characters, and extracts them into structured content — names, contact details, work history, education, and skills.
Step 3: The extracted content is reformatted into your agency's branded template.
The output is a clean, single-column, professionally formatted CV in your agency's style — regardless of how the source was captured.
Step 4: Review and export.
Recruiter reviews the output, applies proofreading if needed, and exports as DOCX or PDF.
Total time: seconds, versus the 30+ minutes it would take to manually retype a scanned CV.
Getting the Best Results From a Scanned CV
OCR accuracy depends on the quality of the source image. A well-lit, straight-on photograph of a clear document will produce near-perfect extraction. A blurry, angled photo of a wrinkled document will produce more errors.
If you're asking candidates to send photos of their CV, a few practical tips help:
- Photo taken in good lighting (natural daylight works well)
- Camera held straight above the document (not at an angle)
- Full page visible in the frame with margins
- Document flat, not curled or folded
- Focus locked on the document before taking the shot
For candidates sending scanned PDFs, higher DPI (300+ preferred) produces better OCR results.
That said, CVFormatter's OCR is trained on the messy, real-world CV images recruiters actually receive — including phone photos with imperfect angles, older scans, and lower-resolution documents. Perfect input isn't required, just readable input.
Note on File Formats
CVFormatter supports the following input formats for scanned or image-based CVs:
- PNG, JPEG, JPG — Image files, processed via OCR
- Scanned PDF — Image-based PDF files (where the "text" is actually rendered as pixels)
- Standard PDF — Text-based PDF files (where text is extractable directly)
- Word DOCX — Standard document files
A quick note: between PDF and Word DOCX, PDF typically produces better parsing accuracy for complex documents. Word files often contain hidden graphics, text boxes, or embedded objects that can confuse parsers. If you have a choice of format from a candidate, PDF is usually the safer request.
For scanned documents specifically, PDF is almost always the format they'll arrive in — but photos (JPEG/PNG/JPG) work equally well.
Quick FAQ
Q: My candidate sent a WhatsApp photo of their CV. Can I use it directly? Yes. CVFormatter accepts JPEG/PNG/JPG image files directly. Upload the photo the same way you'd upload a PDF.
Q: The scanned CV is in a language other than English. CVFormatter's OCR supports multiple languages. If needed, AI translation can be applied as an additional step to convert the extracted content into English (or another target language) for client submission.
Q: The photo of the CV is slightly rotated. Will it still work? Yes, within reason. Minor rotation is corrected automatically. Extreme rotation (upside-down, sideways) may need to be corrected before upload — most phones do this automatically anyway.
Q: Can I anonymise a scanned CV? Yes. CVFormatter's anonymisation works after the OCR step, so candidate names, contact details, and even company names (replaced with accurate industry descriptors based on job title and responsibilities) can all be removed as part of the same workflow.
Q: How accurate is CVFormatter's OCR on scanned CVs? 98%+ across the range of image quality recruiters typically receive. Errors are usually minor formatting issues, not missing content.
Final Thoughts
Scanned CVs aren't going away — they're the norm in many international markets, and they'll continue to arrive from candidates who prefer physical documents or aren't confident using digital tools.
For recruitment agencies that reject them or ask candidates to re-send, the friction adds up. For agencies that can process them directly, it's one fewer thing candidates need to worry about — and one fewer delay in the submission workflow.
CVFormatter turns scanned or photographed CVs into clean, branded, client-ready submissions in seconds. Start your free trial or book a demo to see it work on your own image-based CVs.