Excel to PDF

What a spreadsheet actually stores, and why converters get it wrong

An .xlsx file is a ZIP of XML, and the way it stores a sheet explains most of the bugs you have seen in other converters. Three details in particular do the damage, and all three are handled here.

First, rows are sparse. A row with values in columns A and D lists exactly two cells, each labelled with its own reference. A converter that simply reads cells in order produces a two-column row and shifts every value left — the classic symptom of data appearing under the wrong heading. Cells here are placed by their column letter, so gaps stay gaps.

Second, text is stored separately. Most strings live in a shared-strings table and each cell references one by index, an efficient scheme because a spreadsheet repeats the same words constantly. A reader that ignores that indirection outputs numbers where the words should be.

Third, and most visibly, dates. Excel stores a date as the number of days since 30 December 1899 — 45292 is 1 January 2024 — and records the fact that it is a date only in the cell's number format, in a completely different part of the file. Converters that skip the style table turn every date in your spreadsheet into a five-digit number. This one reads the format table, recognises both the built-in date formats and custom ones, and writes real dates. The odd epoch, incidentally, is a deliberate bug: Lotus 1-2-3 treated 1900 as a leap year, and Excel copied the mistake for compatibility.

How the table is laid out

Each sheet becomes a table starting on a new page, titled with the sheet's own name when the workbook has more than one. Column widths are calculated from the content rather than divided equally, so a column of short reference codes stays narrow and a column of descriptions gets the room it needs. Every column is then scaled to fill the text width of the page.

If you mark the first row as a header, it is drawn bold on a shaded background and repeated at the top of every page the table continues onto — which is the difference between a readable twelve-page table and one where you have to keep flipping back to page one to remember what column four is.

Cell text wraps rather than being cut off, and a row too tall to fit in the space left on a page continues on the next one instead of being clipped. Wide sheets are much better in landscape; the orientation control is right there, and for genuinely wide data a smaller text size buys several more columns.

What this does not try to be

This produces a clean, legible table of your data. It is not a reproduction of Excel's own print layout, and it is worth being clear about the difference. Cell fill colours, borders, conditional formatting, fonts, merged cells, frozen panes, print areas and page-break previews are not reproduced. Charts, pivot tables, images and shapes are not included. Formulas are shown as their last calculated result, which is what the file stores, and macros are ignored entirely.

If what you need is your spreadsheet printed exactly as Excel would print it — with its colours, its merged headers and its carefully set print area — then Excel's own File → Print → Save as PDF is the correct tool, and no browser-based converter is going to match it.

What this is for is the very common case where someone needs the data in a format that opens everywhere, cannot be edited by accident, and does not require the recipient to own a spreadsheet program. A price list to a customer, a timesheet to an accountant, an inventory to a supplier. For that, a clean ruled table is not a compromise — it is usually better than a screenshot of a spreadsheet with gridlines.

Large workbooks, CSV files and old .xls

CSV files are accepted on this page too and converted the same way, which is handy because a CSV opened in Excel and printed is a surprisingly annoying job for something so simple. The delimiter is detected automatically, and quoted fields containing commas and newlines are parsed properly rather than split naively.

The old binary .xls format is not supported: like .doc, it predates the XML packages and is a different structure entirely. Excel, LibreOffice and Google Sheets will all save it as .xlsx for free, and the tool tells you so rather than failing obscurely.

There is no row limit imposed here, but there is a practical one. A sheet of a hundred thousand rows becomes a PDF of a couple of thousand pages, which will take a while to typeset and produce a very large file that nobody will read. If the goal is to hand over a lot of data, a CSV is the better format — and the Excel to CSV tool on this site does that conversion, correctly quoted, in one step.