OCR for Job Seekers: Build a Modern CV From Old Documents

Job hunting in 2026 is a digital sport

If you have not applied for a job in a few years, it can feel overwhelming. Online portals expect a clean Word or PDF document. Recruiters want copy-pasteable skills and education sections. Many companies run an Applicant Tracking System (ATS) that scans your CV electronically and rejects anything it cannot parse.

For most South African job seekers, the problem starts with an old printed CV or certificates that live in a folder somewhere. You cannot edit them. You cannot easily attach them. And retyping the entire document feels like a punishment.

That is where ImageToTextSA fits in.

What you can extract

ImageToTextSA reads the printed text from any clear photograph or scan. Practical examples for job seekers:

• A printed CV from your last job hunt.
• Reference letters from previous employers.
• Matric certificates and university transcripts.
• Skills assessment results.
• Trade qualification certificates and SETA letters.
• ID copies and certified document headers (for your portfolio, not for sharing).

Step-by-step: rebuild your CV in 20 minutes

1. Gather the source material. A clear phone photo or a 300 DPI scan works best. Make sure the lighting is even and the text is in focus.
2. Crop the image to just the document. Phones have a built-in crop tool. Removing the table edges and shadows improves accuracy.
3. Drop the image into ImageToTextSA. Use drag-and-drop or paste from the clipboard with Ctrl/Cmd+V.
4. Press Extract Text. In 5–15 seconds you get a fully editable copy of the content.
5. Clean it up. The OCR engine sometimes confuses "I" with "1" or "O" with "0". Spend two minutes proofreading.
6. Paste into a modern CV template. Word, Google Docs, Pages, or our upcoming free CV Builder will all work.
7. Save as a PDF when you submit — the same PDF download is available right inside the OCR tool.

Build for the ATS, not for the eye

Recruiters spend an average of six seconds on a first scan. ATS software spends zero seconds — it parses text fields. To win both:

• Keep the layout simple: one column, standard fonts (Inter, Calibri, Arial).
• Use clear section headers: Experience, Education, Skills, Languages.
• Put dates on the right of each role, not in tables.
• Avoid icons in headings — they often break ATS parsers.
• Save the final file as a PDF unless the portal explicitly requires .docx.

Common questions

Will my old certificate scan look perfect?

Not always. Old certificates often use ornate fonts that confuse OCR. Expect 85–92% accuracy and budget five minutes to fix the rest by hand.

Can ImageToTextSA read handwritten reference letters?

Neat printed handwriting works. Cursive often fails. If a reference letter is hand-written, ask the referee for a typed copy and OCR the original as a fallback.

Is it safe to upload my ID document?

Your file never leaves your browser. The OCR engine runs locally using WebAssembly. We do not store the image, the extracted text or any personal details.

What if my CV is multiple pages?

Run each page separately, then paste the extracted text into a single CV document. Our PDF download produces a multi-page result, but the OCR step works one image at a time.

What is next for job seekers on ImageToTextSA

We are building a free CV Builder, an Invoice Generator for freelancers, and a PDF → Word converter — all on the same private, in-browser engine. Watch the Coming Soon tools on the homepage.

Good luck with the search. The first step is the hardest, and a clean digital CV is exactly the kind of small win that gets the next application out the door.