How to convert a PDF to Excel

The fastest path. Open a converter that runs in the browser — like the PDF to Excel tool on this site — drop your PDF in, and download a .xlsx or .csv file seconds later.

Because the conversion happens locally via pdf.js, no byte of your file leaves your device. That makes it safe for financial statements, payroll exports, or anything you would not want on someone else's server.

Step-by-step

1. Add files. Drag one or more PDFs onto the converter, use the file picker, or paste from the clipboard (⌘V / Ctrl+V). Up to 20 files, each up to 50 MB.
2. Choose a format. Pick .xlsx for a full Excel workbook with numeric cells, auto-sized columns, and frozen headers — or .csv for a lightweight text file.
3. Adjust options. Toggle “Combine pages” if you want a multi-page statement on a single sheet instead of one worksheet per page.
4. Preview. Expand the preview panel to check the reconstructed table before downloading.
5. Download. Grab individual files or a ZIP of the entire batch.

Microsoft Excel 365 and Excel 2019+ can import PDF data directly via Data → Get Data → From PDF. Excel uses Power Query under the hood: it detects tables on each page and lets you pick which ones to load into the workbook.

LibreOffice Draw can open a PDF, and you can copy-paste table regions into Calc, but the result usually needs manual cleanup because Draw treats each text run as a separate object.

Services like Adobe Acrobat Online, Smallpdf, and iLovePDF upload your PDF to a remote server, run the conversion there, and send back the result. Most offer a limited free tier — typically 1 to 2 files per day — and require a subscription for more.

The upside is OCR: if your PDF is a scan, these services can recognise the text before converting. The downside is privacy — your file sits on a third-party server for at least the duration of the conversion — and the per-file or monthly cost.

When you need to convert hundreds of PDFs or integrate the extraction into a data pipeline, a Python library is the right tool. The three most popular options:

• tabula-py — wraps the Java-based Tabula. Good at stream-mode extraction (whitespace-based) and lattice-mode (ruled-line-based).
• camelot-py — similar dual-mode approach, returns pandas DataFrames directly.
• pdfplumber — pure Python, no Java dependency. Gives fine-grained access to every character and its bounding box.

All three output DataFrames or lists of lists that you can write to .xlsx with openpyxl or pandas.to_excel().

• Check for selectable text. Try highlighting text in your PDF viewer. If you can select individual words, the PDF has a text layer and will convert well. If selecting grabs the whole page as an image, it is a scan and needs OCR first.
• Use “Combine pages” for multi-page tables. Financial statements, invoices, and ledgers often run across several pages. Combining pages puts all rows on one sheet so you do not have to merge worksheets manually.
• Preview before downloading. A quick preview catches problems — merged columns, mis-detected headers — before you build a workflow around the output.
• Prefer .xlsx over .csv for mixed data. The .xlsx format preserves numeric types, percentage formatting, and column widths. CSV is better when you are feeding data into a script or database that expects plain text.

Not every PDF problem is solved by converting to Excel. If the PDF contains mostly prose with an occasional table, extracting just the table (by page range or area selection) will give better results than converting the entire document.

If the data lives in a system that also offers CSV or API export — bank portals, accounting software, CRMs — pull from the source instead. The original data will always be cleaner than a reconstructed table.