Office suites 2027: how to test format fidelity

Nobody buys an office suite for the word processor. Every serious suite can write a letter, add up a column, and put words on a slide. If that were the decision, it would have been settled a decade ago.

What you are really choosing in office suites 2027 is a bundle, a file format, and a set of habits, and of those three, the file format is the one that decides how expensive it will be to change your mind.

What to take away

What you are actually buying

An office suite in this market is rarely three applications. It is a package: documents, spreadsheets, and presentations, plus some combination of mail, calendar, file storage, meetings, and an administrative console to run it all. Microsoft 365, Google Workspace, and Zoho Workplace sell the whole bundle that way, while LibreOffice and ONLYOFFICE concentrate on the editors themselves.

That matters for the comparison in two ways.

First, the price covers the whole bundle. The question is not "is this word processor better" but "is the weakest component in this bundle one I can live with, or will I buy a separate product for it anyway?"

A suite that saves money and forces you to buy a standalone meeting tool has not saved money.

Second, the bundle is where lock-in accumulates. Documents alone are portable enough. Documents plus mail plus storage plus identity plus meetings, all wired together, is an ecosystem, and unpicking it is a project rather than an export.

The five things that actually differ

Strip out the feature lists and the differences that survive are these.

1. Document fidelity across systems. How well a complex document made in one suite survives being opened, edited, and saved in another: a DOCX, XLSX, or PPTX file (the OOXML formats) going into ODF (ODT, ODS, ODP), or the reverse, and how much of it survives the return trip. This is the single most consequential difference and the least discussed.

2. Whether the desktop applications are first-class. Some suites treat the browser as the real product and the installed application as a companion; some do the reverse. Neither is wrong, but the mismatch between the two is felt daily by anyone who works offline, works with very large files, or has years of keyboard habits.

3. The collaboration model. Simultaneous editing by several people, versus one person at a time with the file checked out. Both are legitimate designs and they suit different work.

4. What the administrative layer can do. Sharing defaults, retention, recovery of deleted work, audit logging, and how a departing employee's documents are handled. Invisible in a demo, decisive after two years.

5. Extensibility, and what you have already built on it. Templates, macros, add-ins, and documents that reference other documents. This is the accumulated estate, and it is usually the reason a switch is impossible rather than merely inconvenient.

Format fidelity is the real lock-in

Every suite can open the others' files. That sentence is true and misleading.

Opening a document is not the same as round-tripping it, opening, editing, saving, and having someone on the original system open it again unharmed. Round-tripping is where things go quietly wrong, and the losses are rarely visible on screen:

Two practical consequences. If documents leave your organization for editing (to clients, lawyers, agencies, funders), you are constrained by what they use, not by what you prefer. And if you have a long document estate with heavy formatting, migrating it is a document-by-document exercise, not a bulk conversion.

The test is easy and almost nobody runs it: take your ugliest real document, round-trip it, and look for these six failure points specifically.

This is also the argument for preferring openly specified formats over merely widely used ones. The two that matter are OOXML, published as ISO/IEC 29500, and OpenDocument, published as ISO/IEC 26300; a document in either can be read by more than the vendor that wrote it. The same reasoning sits behind public sector guidance on working with open standards: a format anyone can implement keeps your options open, while a format controlled by one supplier turns a document estate into a reason to renew.

Browser or desktop: the question behind the question

The argument is usually framed as preference. It is really about three concrete situations:

If none of those three applies to your team, the browser-versus-desktop argument is not worth having.

Templates, macros, and the estate you forgot you had

Most organizations carry an uninventoried estate.

It holds a quotation template, a spreadsheet the finance manager built years ago pulling figures from four other spreadsheets, a report that self-formats with a macro, and a mail merge for annual letters.

That estate is the actual cost of changing suite. Before evaluating anything, find it:

You do not need to migrate all of it. But you do need to know it exists before you promise anyone a date.

What to test during a trial

Trials get spent making a new document, which proves nothing. Spend the time on the things that only break under real conditions:

Round-trip your worst existing document, in both directions, and check the six failure points listed above.

Have two people edit the same file at the same time, then have a third join. Watch what happens to formatting while they do it, and check what happens if one of them is offline when they make their changes.

Open your largest spreadsheet and use it properly: sort it, filter it, recalculate it, chart from it.

Delete something and get it back. Find the recovery window, and find out whether recovery restores the file's sharing state as well as its contents.

Share a document with someone outside the organization and have them edit it, using whatever they normally use. This is where fidelity and access controls meet, and it is the scenario people most often discover after committing.

Print something that matters. Pagination differences between suites are real, and if you produce anything that must be a specific number of pages, this is not a small detail.

Look at the admin console for twenty minutes. In Microsoft 365 that means the admin center; in Google Workspace, the Admin console. Find where sharing defaults live, where the audit log is, and what happens to a user's documents when their account is disabled.

What is marketing

Some of the loudest differentiators in this category do not change a buying decision:

Before you commit: the export questions

Ask these while you still have bargaining power:

"Your data is always yours" is a slogan. The answers above are the actual policy.

When you get an answer about formats, sanity check it against a list of formats that institutions consider durable. The National Archives and Records Administration's tables of acceptable file formats published for records transfer are a short, blunt reference. Anything on that list will still open in a decade — PDF/A, the archival version of PDF, is the standard example.

The cost of changing your mind

Not the files. The files move. The costs are:

Which is why the useful moment to think about all this is now, before signing, rather than in three years when the answer is fixed.

Where to read next

Two neighboring decisions are usually made at the same time as this one, and both are settled elsewhere. If the bundle you are considering includes mail, the mail decision deserves its own answer rather than being carried along by the suite, and the hosting arrangement underneath it is a separate question again.

Common questions

Can we run one suite for documents and a different one for mail?

Plenty of organizations do, and it works when the boundary is clear. What it costs you is single sign on across the two, one place to administer people, and the small daily friction of two accounts. Price that friction honestly before deciding it is free.

Is the browser version good enough on its own?

For most people doing most work, yes. It stops being enough for large spreadsheets, heavy layout work, and anyone who works offline regularly. Find out how many of those people you have before choosing a browser only tier, because the answer is usually a small number who matter a lot.

How do we decide if our document estate is too heavy to move?

Take ten documents that would hurt most if they broke, round-trip each one, and count the repairs. Multiply by the number of similar documents you hold. If the number is large, the constraint is real and it should shape the decision rather than being discovered afterwards.

What about the people who will simply refuse to change?

Find out what they actually rely on. It is usually one specific behavior, such as a keyboard habit or a template, rather than the product. Sometimes it can be reproduced and sometimes it cannot, and knowing which before the rollout is worth more than any amount of persuasion.