PDF to Excel

Extracting tables from PDFs is harder than it looks — here's why, and when it works well

PDF tables aren't actually tables in the technical sense. The PDF format doesn't have a table element the way HTML does. What you're looking at in a PDF is a collection of text fragments positioned at specific coordinates, often with visible lines drawn nearby. The converter has to look at those positions and figure out which text fragments belong in which row and column. For a simple, clean PDF exported from Excel or a database report, this works well. For a PDF that was scanned, or one that has tables spanning oddly across columns, or one where someone used tab-spaced text to fake a table visually — it gets messy.

What the output looks like in Excel

Each detected table becomes its own worksheet tab. If a PDF has four pages and two distinct tables, you'll see two tabs in the output .xlsx. Pages with no tables at all still produce a worksheet — it contains the plain-text content from that page rather than leaving you with a blank tab. If a table runs across multiple PDF pages, the rows from all pages are combined into a single continuous block in that worksheet. Repeated header rows that show up on continuation pages are removed so you don't end up with a header row in the middle of your data.

Getting the best results

Use digitally created PDFs where possible — not scanned. Bank statements, invoices, and financial reports exported directly from accounting software convert cleanly. Scanned documents need OCR before extraction will work. Also: the converter reads formula values, not formula expressions, so what ends up in the spreadsheet is what was visible on the PDF page.