A CSV is only text. Turning it into a real spreadsheet means deciding which columns are numbers, dates or codes. We detect the format and suggest a type for every column; switch any of them and see the result.
We converted one source document into every major output format and measured the file that came out. The source was a 37.4 KB DOCX with headings, bold and italic text, a table, numbered and bulleted lists, and one embedded image.
Sizes are from one representative document converted with Pandoc and LibreOffice. Your own files will differ, especially with more images, but the relative ordering holds.
A PDF has no paragraphs. It stores “draw these glyphs at x, y in this font”, often one word or one line at a time, and a scanned page stores only a picture. Drop any PDF to see which kind you have and what its page instructions look like.
European exports use semicolons, databases often use tabs, and some systems use pipes. We test each candidate and pick the one that gives every row the same number of columns. Try reading this file with the wrong one.
,comma;semicolon⇥tab|pipeThese are why people convert: the data stays the same, but it becomes something you can work in.
Log exports and database dumps often run past Excel's limit. Opening them directly silently drops everything after the last row. When a file is too long, we split it across several sheets instead, with the header repeated on each.
The spreadsheet round trip, and where tables come from in the first place.