Most businesses run on spreadsheets, and for a long time that is right. The trouble is the day one of them quietly becomes the system of record.

The spreadsheet that quietly became a system nobody maintains

Somewhere in your business is a spreadsheet that started life as one person's quick calculation and is now load-bearing. Twenty people touch it. The board sees numbers that come from it. It breaks if someone pastes into the wrong cell, and exactly one person understands the whole thing, and that person is overdue some leave. It is not a spreadsheet any more. It has become an unmanaged system, and nobody decided that on purpose.

We cover the practical side of this in our data strategy and leadership work, and rebuilding Wasabi's Microsoft estate looks at a closely related question.

Moving off spreadsheets is not about banning Excel, and any firm that frames it that way earns the resistance it gets. Excel is excellent for a one-off analysis, a model only you use, a scratchpad for thinking. If a handful of people use it, it is not the source of truth, and it would not be a crisis if it broke, leave it well alone. Ripping out healthy spreadsheets to feed a platform nobody asked for is a project that saves nothing and earns resentment.

When it has stopped being a tool

The warning signs are consistent, and hard to unsee once you have seen them. The file gets emailed around and several versions exist with no way to tell which is current. More than a handful of people depend on it. It carries numbers the business reports on. It has grown tabs and formulas only one person understands. There is no record of who changed what, or when, or why. Hit those marks and the risk is no longer inconvenience. It is correctness and continuity.

It helps to name the exposure plainly, because "spreadsheet risk" is too vague to act on. No version control, so people work from stale copies. No audit trail, so an error cannot be traced or a figure defended. Key-person dependency, so capability walks out with one resignation. No real security, so sensitive data sits in a file that can be forwarded anywhere. And silent error, because one mistyped formula can be wrong for months before anyone notices. Each of those is a known way businesses get hurt.

What to keep, what to kill

A move is a chance to leave the cruft behind, not to recreate every tab faithfully in a more expensive tool. Keep the genuine analysis and the human judgement; those belong with people and always will. Move the repeatable, shared, source-of-truth work onto a platform where it can be governed, refreshed automatically, secured and trusted. Kill the duplicated, the abandoned, and the reports nobody reads. Migrate everything exactly as it was and you have not modernised, you have moved the mess into a smarter building.

Start with the worst one, not the easiest

You do not need a grand programme to begin. You need one painful spreadsheet replaced properly, end to end. Pick the one whose failure would hurt most, not the one that looks simplest. Get its inputs flowing in automatically instead of being pasted each cycle. Rebuild its logic as agreed definitions in a small model, not a maze of formulas only one person can read. Then run the new version alongside the old until the numbers agree and people trust it, and only then retire the file.

One spreadsheet done properly teaches you more, and builds more confidence across the business, than a year of planning. The second is faster, the third faster still, and the habit spreads on its own once people see it work.

The opposite mistake is just as expensive, and easier to make with a budget in hand. A business that closes its books in a morning does not need a vast lakehouse to replace a spreadsheet. Match the platform to the size of the problem. A trusted, maintainable number is the goal, not sophistication.

So this week, name your one load-bearing spreadsheet, the one whose failure would genuinely hurt, and ask who would rebuild it if its author left tomorrow. If the answer is nobody, you have found where to start. That first move is exactly what the Establish phase of our Analytics Acceleration Programme and our data platform work is built to do, and the full ninety-day version is in the Off the Spreadsheets guide.

If any of this sounds familiar, talk to us about your data.

Related reading

• Analytics Is A Garden, Not A Project
• Consolidating multiple Business Central companies in one model
• Why Your Self-Service BI Rollout Collapsed