Convert PDF to Excel

Real cells. Each value goes into its own cell, and anything that reads as a number is stored as a number rather than as text that looks like one — which is the difference between a column you can sum and a column you cannot.

By its columns. A table draws its cells at repeating left edges, so a handful of positions each shared by several rows is strong evidence of one. Prose has no such regularity and produces a single column, which is correctly read as a sign that this is not a table.

You are told, and no file is produced. Turning a page of paragraphs into rows and columns would give you a spreadsheet full of fragments; refusing is more useful. The message points you at PDF to Word instead.

Yes. Each page that contains a table becomes its own worksheet, named for the page it came from, so a report with a table on pages three and seven produces two sheets.

Not reliably. Column detection works from left edges, so a cell whose text wraps onto a second line, or one that spans several columns, may land in the wrong place. Check the result against the original before relying on the figures.

No. A scan holds no text, only a picture of one, so there is nothing to place into cells. It will tell you rather than producing an empty workbook.

Neither, and neither exists in the source. A PDF stores the result of a calculation, not the calculation, so what you get is the values. Colours, borders and fonts are not carried across.

Any spreadsheet application — Excel, Numbers, Google Sheets, LibreOffice Calc. The file is a standard .xlsx workbook.

They do not. Finding the tables and writing the workbook both happen inside this tab, which matters more here than almost anywhere else on the site — tabular PDFs are usually invoices, payroll and bank statements. Nothing carrying the document goes out, and it works with the network switched off entirely.