Office alternatives: "opens .docx" is not the question

Every comparison of Microsoft Office alternatives leads with a compatibility checkmark. Opens .docx. Opens .xlsx. Saves back to both.

That checkmark is true and it is not the question. Opening a Word file is solved and has been for years. What is not solved is what happens on the second trip — when your edited document goes back to a colleague running real Office, they change something, and it comes back to you again.

That round trip is where switching succeeds or fails, and no feature table describes it.

Why the format standard does not settle it

The .docx format is an international standard, ISO/IEC 29500. That sounds like it should end the argument, and it does not, for a reason worth understanding before you choose anything.

The standard defines two conformance classes. Strict is the clean specification. Transitional carries a large catalogue of compatibility features, deprecated elements and platform-specific behaviours, included so the standard could faithfully represent documents produced by Microsoft Office going back to the 1990s.

The practically important fact: Transitional is the variant in widespread use. Documents you receive are overwhelmingly Transitional, which means "supports ISO/IEC 29500" tells you much less than it appears to. Free-software advocates and the Document Foundation have made this point pointedly — specifying the standard in a procurement document does not, by itself, buy interoperability.

The consequence for you is simple. Compatibility with .docx is not a yes or a no. It is a percentage, and the percentage depends on which features your documents actually use.

The four things that break, roughly in order

Track changes and comments. The most common real failure. Revision marks and threaded comments are the features that survive a round trip least reliably, and they are precisely what you use when a document is moving between people. A document you only read will be fine. A document you negotiate over is the test case.

Complex layout. Text boxes, floating images with tight wrapping, multi-column sections, and anything anchored to a specific position. Simple documents port cleanly; designed documents drift.

Macros. VBA does not port. Alternatives have their own scripting, which is a different language and a rewrite, not a conversion. If your spreadsheet runs on macros, this is a project and not a switch.

Fonts, which people underestimate. Microsoft's interface fonts are licensed, not free to redistribute, so alternatives substitute a lookalike. A substituted font has different metrics — characters occupy different widths, so lines wrap in different places, so page counts change and page breaks move.

That last one got sharper in January 2024, when Microsoft replaced Calibri as the Office default after seventeen years with a new typeface called Aptos. Any document created under the new default carries a font your alternative probably does not have, and the drift shows up as pagination, which is exactly what people notice in a printed contract.

The round-trip test

Here is the test that settles it for your documents, rather than for documents in general. It takes about an hour and needs one cooperative colleague running real Office.

1. Take a genuine document of yours — the most complicated one you regularly handle. Not a sample.
2. Open it in the alternative you are evaluating. Make three kinds of edit: normal text, a tracked change, and a comment.
3. Save as .docx and send it to your colleague on real Office.
4. Ask them to accept one of your tracked changes, reject another, reply to your comment, and send it back.
5. Open it again in the alternative.

Then check, in this order: did the tracked changes survive as tracked changes, or did they get flattened into plain text? Is the comment thread intact with its replies attached to the right anchor? Did the page count change? Did anything move?

Flattened revision marks are the failure that matters most, because it is silent. Nothing warns you. The document simply looks finished when it is not, and the record of who changed what is gone.

Run the same test with a spreadsheet if you use spreadsheets seriously. Formulas port well; conditional formatting, pivot tables and charts are the ones to watch.

Reading the result honestly

Everything survived. Switch. Your documents are within the well-supported subset, which for a lot of people is genuinely the case.

Layout drifted but content and revisions survived. Switch for drafting and keep one Office licence for final formatting. This is a common and sensible end state, and it costs far less than licensing everyone.

Revision marks or comments broke. Do not switch for documents that move between people. You may still switch for the documents that do not.

Macros are involved. Treat it as a migration project with a budget, not a software choice.

What actually differentiates the alternatives

Once you know how much fidelity you need, the remaining differences are real:

• Interface familiarity. Some alternatives copy the Office ribbon closely; others follow a different tradition. For a team that is switching under mild protest, this matters more than any feature.
• Where the file lives. Desktop application with local files, browser application with cloud files, or both. This changes your backup story and your privacy exposure, not just your convenience.
• Licence model. Free and open source, one-time purchase, or subscription. A one-time purchase that stops receiving compatibility updates ages badly in this specific category, because the thing it must stay compatible with keeps changing.
• Whether it bundles things you did not ask for. Read the installer screens. Some free office suites monetise through bundled extras or aggressive upgrade prompting, and that is part of the price.

The recommendation we are comfortable making

If your documents rarely leave your own hands, the free options are genuinely sufficient and we would start there rather than pay. LibreOffice costs nothing and is not a trial.

If documents circulate with tracked changes, run the round-trip test before you commit anything — including before you commit to a paid alternative, because paying does not guarantee fidelity either.

If you are switching to save money on a team, price the hybrid honestly: alternatives for most people, a small number of real Office licences for whoever produces the final document. That is usually cheaper than either extreme and it removes the failure mode that causes teams to switch back.

What we have not tested

We have not run the round-trip test across the current versions of the major alternatives and published the results. That is the article we would rather have written, and it requires methodical work with a controlled document set and a second machine running current Office.

Everything above is drawn from the format standards, the Library of Congress format registry, and published documentation on OOXML conformance. The description of what breaks reflects well-documented characteristics of the format boundary, not measurements we have taken ourselves.

We will publish the measured version when we have it, and we will say so here if the numbers contradict anything above.