Turn PDF bank statements into clean CSV or Excel. The file is parsed inside your browser — it is never uploaded to any server, because there is no server.
or click to choose files — multiple statements are merged into one sheet
Processed locally
| Date | Description | Reference | Debit | Credit | Balance |
|---|
A bank statement converter takes a PDF that a bank generates for your account and pulls the transaction rows out of it into a spreadsheet. Banks export PDFs for humans to read, not for software to parse: the columns shift between pages, the dates use a dozen different formats, and running totals appear mid-table. A converter reads those table structures and produces one clean row per transaction with a date, a description, an amount and a running balance.
Accountants do this by hand, copying row by row from the PDF into Excel. A single month of a busy business account can be 300+ lines. That is the hours this tool is built to remove.
Your PDF is read with a JavaScript PDF library inside this page. Text is extracted per line, then transaction rows are identified by finding a date and a monetary amount on the same line. Amounts are classified as debit or credit by position and sign, and the running balance is read when present.
No template matching. It does not require one specific bank layout, because banks all format statements differently. Rows are found by structure rather than by a fixed template, so the same parser handles statements from different banks.
Most online converters upload your PDF to a server, process it there, and hold a copy you cannot see or delete. For a bank statement that is a real problem: the document contains your account number, your balance, your salary credits and every merchant you spend at.
This one never sends the file anywhere. The parsing runs in your browser's JavaScript engine, so the bytes go from your disk to your own CPU and nowhere else. You can verify it: disconnect from the internet after the page loads and it still works.
Scanned image-only PDFs contain no text layer, so they cannot be parsed by any browser-based converter. If your rows do not appear, the statement is almost certainly a scan — OCR it first.
No. The PDF is parsed by JavaScript running in your browser. No network request carries your file. You can confirm this by opening your browser's network inspector while converting, or by going offline after the page loads.
Almost always because the statement is a scanned image with no text layer. Browser-based tools cannot read those. Run the PDF through OCR first, then convert it.
Parsing is heuristic, so check the preview before importing. The preview shows every extracted row. Any row that looks wrong is visible immediately rather than silently corrupting your spreadsheet.
It is not template-based, so it handles most banks that produce text PDFs. Some banks use unusual table structures or split dates across lines, and those may extract partially.
Yes. Select several PDFs and they are merged into a single sheet, sorted by date.
Browser memory is the limit, not the server. Statements up to a few hundred pages work fine. Very large scanned documents can exhaust memory in the tab.