When has a spreadsheet become a business risk?

A spreadsheet becomes a business risk the moment losing it, or losing the one person who understands it, would genuinely hurt. Until then it's a brilliant, cheap, flexible tool and you should keep using it. The skill is noticing when it has quietly crossed that line.

Spreadsheets don't fail loudly. They fail on the day someone is on leave, or a formula silently breaks, or two people edit two copies and nobody knows which is real. Here are the signs the sheet has outgrown its job.

Five signs to watch for

• Only one person truly understands it. If your operations depend on a single colleague's spreadsheet, you don't have a system, you have a dependency.
• There are several copies, and no one is certain which is current. The moment the words 'final_v3_latest' appear, the sheet is managing you.
• People are re-typing its numbers into other tools. Manual re-entry is both a time cost and a steady source of quiet errors.
• It has become a database, not a calculation. Hundreds of rows of live records, customers, orders, stock, is a job a spreadsheet was never built to hold safely.
• A wrong cell would be expensive. If one bad edit could misprice an order or lose a record, the risk now outweighs the convenience.

None of these mean 'panic and buy enterprise software.' They mean the sheet has taken on a load it isn't designed to carry, and it's worth planning the next step deliberately rather than after something breaks.

What the sheet was always good at

It's worth being fair to the spreadsheet. It's unbeatable for thinking, modelling, one-off analysis, and prototyping a process before you know its final shape. Many good custom tools begin life as a spreadsheet, that's how you learn what the process actually is. The problem is never that you used a spreadsheet. It's that the process quietly graduated and the tool didn't.

What to do about it

You don't have to replace everything. Usually the right move is to lift out the one part that has become risky, the shared list of live records, the critical calculation, the step three people re-type, and give it a proper home: a small internal tool with real access control, a single source of truth, and a history of who changed what. The rest of the spreadsheet can happily stay a spreadsheet.

The test for whether it's worth building something is simple: would a mistake in this sheet, or the loss of the person who runs it, cost you more than fixing it would? When the answer becomes yes, it's time, and not a moment before.