Convert PDF to CSV in Your Browser
Extract tables from a text-based PDF and download them as CSV, entirely in your browser. Your PDF never leaves your device, and the tool keeps working offline.
Drag and drop or Tap to select a PDFclick to select a PDF
Files are processed on your device only.
How it works
- Drop or select your .pdf file. It is read on your device only.
- Columns are reconstructed and shown as CSV in a preview.
- Click Download to save the .csv file.
How tables are reconstructed from a PDF
A PDF does not store a table as a grid; it stores each piece of text at an x and y position on the page. To rebuild a table, the tool groups text into rows by vertical position, then infers the columns by clustering the horizontal positions where text starts, and finally places each value into the column it lines up with.
That means the accuracy depends on the spacing in the original file rather than on a hidden structure. To turn CSV data back into a printable table, the CSV to PDF converter goes the other way. See how browser-side rendering works.
What works well, and what to double-check
Clean results come from tables with a single header row, consistent columns, and visible gaps between them, the kind exported from a spreadsheet or a report generator. The header and each data row map straight to CSV lines.
Check the output when a table has merged or multi-line cells, tightly packed columns, or several tables on one page, since column inference can drift. If you only need the raw words rather than a grid, the PDF to TXT page extracts plain text instead.
Frequently asked questions
- Does this read tables from scanned PDFs, and is the result exact?
- No to scans: a scanned or photographed table is an image with no text to read, so it comes back empty. For text-based PDFs the columns are reconstructed from where the text sits on the page, which is a best-effort estimate: simple grids come out clean, but complex tables may need a quick fix-up.
- Which tables convert cleanly?
- Plain single-header grids with clear column gaps convert well. Merged cells, nested headers, cells that wrap onto several lines, and columns that are very close together are the cases most likely to need review, because the tool infers columns from spacing rather than from a stored grid.
- What does the downloaded file look like?
- It is standard comma-separated values: one line per detected row, cells joined by commas, and any cell that contains a comma, quote, or line break is quoted. You can open it in a spreadsheet or feed it to a script directly.
- Is my PDF uploaded anywhere?
- No. The PDF is opened and read in your browser on your device with a local engine. Nothing is sent to a server, there is no account, and it keeps working offline.