Extract data from a scanned paystub
Scanned PDFs are just pictures of text, so copy and paste doesn't work. StubParse runs OCR on each page, or sends up to five pages to AI mode, and returns the fields as structured data.
Start free3 documents free · No card required
Multi-page scans
AI mode reads up to five pages of one stub together and combines them into a single result.
Structured, not just text
You get fields like gross pay, each tax, and each deduction, not a wall of OCR text.
See the raw text too
The extracted text is available under each stub if you want to check what OCR saw.
How it works
- Step 1
Drop the stubs
PDFs or photos, one or a batch. Files are read in your browser.
- Step 2
Review the numbers
Every field is editable, and three math checks re-run as you type.
- Step 3
Export
Excel with styled headers, a flat CSV, or JSON. Include YTD or leave it out.
Questions
- Why can't I copy text from my scanned PDF?
- A scan stores each page as an image. OCR is needed to turn the image back into text.
- How many pages can it read?
- Text PDFs are read in full. Image-only scans go to AI mode up to five pages per stub.
- What if the scan is low quality?
- Try AI mode. If numbers still come out wrong, edit them before export; the math checks point to mismatches.