How to convert a scanned bank statement to Excel (OCR explained)
If your statement is a scan or a photo, getting it into Excel is harder — and riskier — than it looks. Here's why, and the most accurate path.
Text PDF vs. scanned PDF
There are two kinds of "PDF":
- Text PDF — generated by your bank's system. The numbers are real, selectable text. These convert cleanly and accurately.
- Scanned PDF — a photo or scan of a printed page. It's just an image; there's no real text underneath, so a computer has to guess the characters with OCR (optical character recognition).
Quick test: open the PDF and try to select a number with your mouse. If you can highlight it, it's a text PDF. If you can't (it selects like an image), it's scanned.
Why OCR is risky for money
OCR reads pixels and predicts characters, so it makes classic mistakes on financial data: a "5" read as "S", a "1" as "l", a decimal point dropped, or two cramped lines merged. On a column of numbers, a single misread digit is a wrong amount you might not catch. That's why OCR'd statements always need careful checking.
The most accurate path: get a text PDF
Before reaching for OCR, try to get the original text PDF:
- Log into your bank's website or app and download the statement directly (almost always a text PDF) rather than scanning a paper copy.
- Ask your bank to re-send an electronic statement if you only have paper.
Once you have a text PDF, LINGYANG converts it in your browser with no OCR guessing and no upload — the numbers come through exactly as printed, and you review them before exporting.
Got a text PDF from your bank? Convert it free, in your browser.
Open the converter →If you only have a scan
LINGYANG currently works with text PDFs, not images, so a pure scan won't extract here yet. Your options: (1) get the original text PDF from your bank (best), or (2) use a dedicated OCR tool and then carefully verify every amount against the scan before trusting it. Either way, double-check the totals.
Have a text PDF instead? Convert it now →
Key takeaways
OCR is useful but risky for money. Use a text PDF when you can; when you only have a scan, choose clear images and review every amount.
- If you cannot select text in the PDF, it is probably scanned.
- OCR can confuse digits, punctuation and column boundaries.
- Image quality, contrast and straight pages matter more than file size.
- Review against the original statement before exporting or importing.
Before you export
- Prefer a text-based PDF downloaded directly from online banking.
- Review dates, descriptions, signs, amounts and balances against the original statement.
- Export a small sample first if you plan to import into accounting software.
- Keep statement contents in Local Mode when privacy, client confidentiality or compliance matters.
Frequently asked questions
How do I know if a PDF is scanned?
Try selecting a transaction line. If you can only select the whole page image, it is scanned and needs OCR.
Is OCR accurate enough for accounting?
It can be, but only after review. Money data should never be used without checking amounts and dates.
Does LINGYANG use cloud OCR today?
No. Current OCR runs on device with self-hosted assets. Cloud OCR is not enabled today.
Next step: OCR is useful but risky for money. Use a text PDF when you can; when you only have a scan, choose clear images and review every amount.
Learn about local OCR →