What to do next
Each sign marks a specific structural limit, and naming the limit tells you what to replace the spreadsheet with — because "get off spreadsheets" is useless advice without a destination. Version confusion (sign one) means you need a single source of truth: a shared database or the record system inside an industry platform, where there is only ever one copy of a customer. Multiplayer collisions (sign two) mean the same, plus permissions. Deadline tracking (sign three) is the most dangerous one, because a spreadsheet cannot act — it will hold "follow up March 14" in a cell forever without ever telling you it's March 14. Anything time-triggered belongs in a system that fires reminders and sequences, not one that stores text. The costly-typo sign means you need validation — software that refuses an impossible date or a malformed price — and the report-assembly sign means your data should live where reporting is a button, not a Sunday ritual.
Be equally clear about the false signs, because spreadsheet guilt drives bad purchases. Having many spreadsheets isn't the problem — a stable, single-purpose sheet that one person updates and no deadline depends on is a perfectly good tool, and migrating it to a $99-a-month platform is a downgrade wearing a demo. The failure conditions are shared editing, time-triggered actions, and scale — not spreadsheet-ness itself. Plenty of solid businesses run a hybrid: a real system for customers, jobs, and money; spreadsheets for the odd analyses at the edges. That hybrid is usually the right end state, not a way station.
The migration itself is where owners stall — years of data, mid-flight operations, no obvious moment to switch. The unstuck version: migrate one workflow, not everything; run the new system for new records while the old sheet stays read-only history; and pick the workflow where the spreadsheet failure is costing the most, which is almost always the deadline-dependent one. Knowing which workflow that is — priced, not guessed — is the kind of finding a FrictionList audit is built to produce: several audit clients have arrived asking "which platform should we buy?" and left with a cheaper answer, because the expensive leak was in a different spreadsheet than the one they suspected.
Quick answers
When is a spreadsheet no longer enough for a small business?
When multiple people edit the same data, when deadlines depend on it, or when errors in it have real financial cost — any one of these is the threshold.
What should replace a spreadsheet first?
The deadline-dependent workflow — follow-ups, renewals, jobs in progress — because spreadsheets store dates but can't act on them.
Are spreadsheets ever fine to keep?
Yes: single-owner, no-deadline, low-stakes uses like ad-hoc analysis. The failure conditions are sharing, timing, and scale — not spreadsheets themselves.