
There's a spreadsheet in every company that runs something critical. It might be the dispatch schedule, the client onboarding tracker, the commission calculator, or the inventory reconciliation sheet. It started simple - someone needed to track something - and grew until it became the operating system for an entire workflow.
These spreadsheets are simultaneously the most important and most fragile part of the operation. They're important because the workflow doesn't function without them. They're fragile because they depend on one person who understands the formulas, break when someone edits the wrong cell, and can't handle concurrent access, versioning, or audit trails.
Why Spreadsheets Persist
Spreadsheets persist because they solve the problem immediately. When a new workflow emerges, nobody files a ticket with IT and waits three months for a system. Someone opens Excel and starts building. Within a day, there's a working solution.
This is genuinely valuable. Spreadsheets are the fastest prototyping tool ever created. The problem isn't that people use spreadsheets - it's that they never graduate from them. The prototype becomes the production system by default, not by design.
The graduation doesn't happen because the spreadsheet "works." It handles the current volume. The person who built it maintains it. The issues - broken formulas, version conflicts, no audit trail - are absorbed as normal operating friction. By the time the spreadsheet's limitations become critical, it's so embedded in the workflow that replacing it feels impossible.
The Signals It's Time
Several signals indicate a spreadsheet has become a system requirement:
Multiple people need to update it simultaneously, and they're working around conflicts with "don't touch the sheet between 2-4 PM" rules.
The spreadsheet has more than 10 tabs, or the formulas reference across sheets in ways that one person understands.
A new hire needs more than a day of training to use the spreadsheet correctly.
Someone has written macros or VBA to extend the spreadsheet's functionality.
The team has a "backup person" who can maintain the spreadsheet if the primary owner is unavailable.
Any one of these is a signal. Two or more is urgent. The workflow has outgrown its tool, and the gap between what the spreadsheet can do and what the operation needs is widening daily.
Several signals indicate a spreadsheet has become a system requirement: Multiple people need to update it simultaneously, and they're working around conflicts with "don't touch the sheet between 2-4 PM" rules.
The Migration Pattern
Moving from spreadsheet to system doesn't require a big-bang replacement. The pattern that works:
First, formalize the data model. The spreadsheet's columns are an implicit schema. Make it explicit. Define the entities, relationships, and constraints. This usually reveals inconsistencies in the spreadsheet that have been causing errors nobody noticed.
Second, codify the logic. The formulas and manual steps are an implicit workflow. Write them as explicit business rules. This usually reveals undocumented decisions that live in the operator's head.
Third, build the minimum system. Not a feature-complete replacement - a system that handles the core workflow with proper data integrity, concurrent access, and audit trails. Import the existing data. Run both systems in parallel for a week.
Fourth, extend. Now that the foundation is solid, add the capabilities the spreadsheet couldn't provide: automated triggers, integrations with other systems, reporting that isn't a pivot table, role-based access that isn't "don't edit this tab."
What Gets Unlocked
The transformation from spreadsheet to system typically unlocks capabilities the team didn't know they were missing. Real-time visibility instead of stale snapshots. Automated notifications instead of "remember to check the sheet." Integrations with upstream and downstream systems that eliminate manual data transfer. Historical analytics that go beyond what's in the current file.
But the biggest unlock is resilience. The workflow no longer depends on one person, one file, or one machine. It runs on infrastructure that handles backups, access control, and concurrent use. The operations team stops worrying about the spreadsheet and starts focusing on the work the spreadsheet was supposed to support.
If your company has a spreadsheet that more than three people depend on for daily operations, that's not a spreadsheet. It's a system specification written in Excel. Treat it that way, and build accordingly.
Related

Why Copilots Miss the Point
Copilots keep humans doing the wrong work. The goal isn't better tools for manual tasks - it's eliminating the manual tasks entirely.

Anatomy of an Autonomous Operation
What it actually means for a system to operate autonomously. Decision loops, exception handling, state management, self-correction.
Building something that should run itself?
Start Building