When spreadsheets stop scaling: a five-signal test
Quick answer: Five signals tell you a spreadsheet has stopped serving the business: regular data integrity errors, multi-user conflicts, no audit trail, an integration ceiling that forces manual data copying, and only one person who actually understands how it works. If two or more apply, the cost of staying is now larger than the cost of moving.
Most 30 to 200 staff businesses we work with are running at least one critical operation on a spreadsheet that's been around for years. The spreadsheet started as a quick fix. Then it grew. Then it became "the spreadsheet". Now it has six tabs, three macros, a sheet called "DO NOT EDIT" that everyone edits, and a filename that ends in _FINAL_v3_REAL.xlsx.
If you're reading this, you probably know exactly which spreadsheet I'm describing in your business. The question is whether you've crossed the line where keeping it costs more than replacing it. Here's a five-signal diagnostic to help you decide, plus a few honest words about what staying is actually costing you.
Why do so many SMEs still run their business on spreadsheets?
Spreadsheets are remarkable software. They're flexible, they're cheap, they're well-understood, and almost every Australian SME has people who can edit them confidently without training. For most operational work, they're a reasonable choice. The fact is, a lot of business runs on Excel and Google Sheets, and that's not the problem we're here to solve.
The problem is that spreadsheets that solved a real operational problem in 2018 keep running for years after the business has changed shape around them. The team gets bigger. More people need to use the file at the same time. The data gets connected to other systems that didn't exist when the spreadsheet was built. The original creator leaves the business. The work the spreadsheet does becomes more important than the people running it realise. The spreadsheet has been outgrown without anyone noticing the moment it happened.
There's also a separate category where spreadsheets aren't an option in the first place because the calculation is genuinely outside what they can do. Geospatial analysis is the clearest example. Our own Nuonic platform exists because matching millions of GPS points against road network polygons to identify off-road operation for Fuel Tax Credits is not something a spreadsheet can do at any size. Everything else around the FTC process could be handled in Excel; that one calculation required complex, data-intensive code. Nuonic abstracted it away and made the result usable by small operators who otherwise wouldn't have had the capability. That's a different kind of mismatch than the one this article is about. We're focused here on the more common case: spreadsheets that worked in principle and have stopped serving the business in practice.
How do I tell if my spreadsheet has stopped serving the business?
Run these against the spreadsheet you have in mind. If two or more apply, the cost of staying is now larger than the cost of moving.
1. Data integrity errors are happening regularly.
Formulas break when someone inserts a row in the wrong place. Conditional formatting stops working. Someone overtypes a formula with a value and nobody notices for a week. The numbers occasionally come out wrong and someone has to investigate. If you've started building "checks" for the spreadsheet's output, the spreadsheet itself is no longer trustworthy.
2. Multi-user conflicts are part of life.
Two people are editing at the same time. One person's changes overwrite another's. Someone has to coordinate who's "in" the file. The cloud version tries to handle this and almost manages but not quite. The version control is happening through filenames that look like quotes_FINAL_v3_REAL_USE_THIS_ONE.xlsx. You laugh about it, but it's a real problem.
3. There's no audit trail.
When something goes wrong (a wrong number sent to a customer, a missed deadline, a regulatory issue), nobody can answer "who changed this and when?" The spreadsheet shows the current state, not the history. For anything operationally important, that's a real risk, not just an inconvenience.
4. The spreadsheet has hit an integration ceiling.
Data has to be manually copied between the spreadsheet and your other systems. Someone exports from the CRM, pastes into the spreadsheet, runs the formulas, and pastes the results back. The "integration" is two staff with copy-paste fatigue. If the manual reconciliation work is more than a few hours a week, the spreadsheet is now a bottleneck.
5. Only one person actually understands how it works.
Sarah built the original version and has been the keeper ever since. The macros and formulas have grown organically. There's no documentation. If Sarah leaves, you have a real problem on your hands. This is the most under-counted risk in SME operations and the one most likely to surface as a crisis rather than a slow degradation.
What does it cost to keep using a stretched spreadsheet?
The hidden cost of staying on a stretched spreadsheet shows up in three places.
The first is time. In our experience, the businesses we observe locally typically have two staff spending close to half their time keeping the spreadsheet aligned with reality, dealing with errors, reconciling between systems, and answering "is this number right?" questions from people downstream. That's a full FTE of operational overhead nobody is counting because it's distributed across two part-time burdens.
The second is risk. Errors that get caught are an annoyance. Errors that don't get caught reach customers, regulators, or the bottom line. The audit trail problem and the integration ceiling problem both make the cost of an undetected error larger over time, not smaller.
The third is opportunity cost. Every hour spent maintaining the spreadsheet is an hour not spent on whatever the business actually competes on. For businesses with thin margins, that compounds quickly.
What is clear is that the cost of moving has come down. AI-assisted custom software development has reduced the cost of replacing an SME-scale operational spreadsheet with a fit-for-purpose application by close to half over the last 18 months. We've written about the broader build-or-buy calculation and what custom software actually costs in 2026 for the depth. The point here is simpler. The cost of staying with a stretched spreadsheet hasn't moved while the cost of fixing it has dropped. The math has changed.
What to do next
Most spreadsheets that have stopped serving the business will keep running for years before they finally fall over. The cost of staying compounds quietly. The cost of moving has dropped. In our experience working with businesses across Australia from our base in Brisbane, the right time to move is usually 12 months earlier than it feels.
If you've got a critical spreadsheet you've been meaning to replace and you want to test the math on what it would actually cost, that's exactly what our Discover phase is for. Two to four weeks, prioritised, costed, no strategy decks. Start a conversation.
Frequently asked questions
Is it cheaper to replace a spreadsheet with custom software or with SaaS?
It depends on whether the SaaS actually fits how you operate. If something like Monday.com, HubSpot, or Xero matches how your team works today, buy SaaS. If you'd be reshaping your operations to fit the tool, custom software is usually the better long-term answer. The build-or-buy calculation has shifted in the AI era; we've covered it in detail here.
How long does it take to move a critical operation off a spreadsheet?
For an SME-scale spreadsheet replacement, three to five months is typical in 2026, depending on the integrations involved. The build work is usually less daunting than people fear. The transition (running both systems in parallel for a few weeks, training users, sorting out edge cases) is the part most operators underestimate.
We've been talking about replacing this spreadsheet for years and never have. Why does it keep getting deferred?
Because the spreadsheet still works most of the time, and the business has bigger fires to put out. The cost of staying is real but distributed across many small moments, while the cost of moving feels concentrated and risky. The deferral is rational in the short term and expensive in the long term.