In defence of the Google Sheet: it is, genuinely, the right first tool for most growing businesses. Free, universally understood, and it stretches to cover a surprising amount of business logic before complaining. I’ve seen real operations run on a single well-structured sheet for years.
But at some point, quietly, the sheet stops serving the business and starts slowing it down. Most founders don’t notice when this happens — they notice three months later, when something important goes wrong. Here’s a diagnostic for catching the moment earlier.
The seven signs
1. Two people editing the same rows keep stepping on each other. Google Sheets handles concurrent edits gracefully for small numbers of cells. The moment ops wants to bulk-update 200 rows while finance is reconciling in there, you get the polite but deadly “Someone else is editing this” dialog — and real work stops while people take turns.
2. You have a formulas guy. One person actually understands the VLOOKUPs, the nested IFs, and the tab on tab C that nobody is supposed to touch. When they go on leave, the business slows. When they leave the company, the sheet becomes a sacred object nobody dares maintain — and the knowledge walks out the door with them.
3. Reports require manual copy-paste between tabs. You can spot this when someone is spending Friday afternoons “preparing the weekly numbers.” What they’re actually doing is hand-reconciling data that should be moving itself. The reports get stale and error-prone in proportion to the size of the manual step.
4. You can’t answer “how many X did we do last month” without a twenty-minute session. The sheet holds the data but can’t answer the question. You filter, then pivot, then copy into another sheet. The friction means you stop asking — and decisions end up being made on gut feel when they should be made on data you already have.
5. Data entry has drifted. Two people enter the same thing in different formats. “Kuala Lumpur”, “KL”, “Kuala Lumpur ”. “Pending”, “pending”, “PEND”. You know there are duplicates. You’re not entirely sure how many. You’re definitely not cleaning them up this week.
6. You’re afraid to touch it. You’ve been meaning to fix one thing for weeks but you’re worried you’ll break something else. When a tool makes you afraid of it, it’s no longer a tool — it’s a liability you’re managing around.
7. The non-technical people avoid it entirely. They use the sheet through someone else. They Slack the “formulas guy” instead of opening the file. They keep their own side-sheets. The central thing has become central only nominally.
Three or more of these and the sheet has already lost. You’re just living in the lag between the moment it stopped being useful and the moment you admit it.
What “replacing it” actually looks like
Most founders, when they finally concede, imagine a big-bang rebuild: a project, a launch, a migration weekend, a “new system” announced to the team. That’s almost always the wrong shape — expensive, slow, and it fails the moment anyone finds something the old sheet did that the new system doesn’t.
The right shape is smaller: pick the one workflow that hurts most and build a focused web tool for that. Not the whole business. Not every sheet. Just the bit that’s costing you the most time or the most trust. Move the team over to it for that one workflow. Leave everything else on the sheet. Then repeat.
Over six to twelve months, the sheet gradually shrinks back to what it was good for in the first place — ad-hoc analysis, one-off calculations, exploratory work. The operational workflows move out, one at a time, into tools designed for them.
What to keep the sheet for, forever
Exploration. Prototyping a workflow before you commit to building it. Sending data to your accountant. Anything that happens once. Anything where the audience is one person.
The spreadsheet was never the problem. The spreadsheet-doing-the-job-of-an-internal-tool was the problem. Once you can tell the difference, the path forward stops being a big migration and becomes what it should have been all along: a sequence of small, unglamorous decisions, each one replacing one workflow at a time.
That’s a project you can actually finish.