How Excel becomes your operating system
Excel is usually the first tool people reach for when they need to calculate, compare, or track something. It’s universal, accessible, and easy to use, so anyone can start working with it immediately. Packy McCormick called it “probably the most influential piece of software ever built,” and it’s hard to disagree. For many business uses, it works remarkably well.
The problem starts when processes that began in a spreadsheet stay there long after their complexity outgrew what the tool can handle. A sheet for tracking orders becomes an order management system used across departments. Next to it grows a contractor billing file and a resource planning sheet — each built for the moment, for a specific need. People reach for Excel to get their work done efficiently, and over time their individual sheets add up to a patchwork system the entire organization ends up depending on.
What drives up the real cost of Excel
In 2003, Canadian energy company TransAlta lost $24 million — 10% of its annual profit — because of a spreadsheet error. High bids were mapped to the wrong contracts, and the company overpaid on every single one. CEO Steve Snyder publicly confirmed it was “literally a cut-and-paste error in an Excel spreadsheet.”
That’s a spectacular example, but the mechanism behind it is common. The difference is that losses spread across hundreds of small errors that nobody ever counts.
According to Raymond Panko’s research, conducted over more than 20 years, 88–94% of operational spreadsheets contain errors. 82% of those are logic errors — not typos, but mistakes baked into the structure: incorrect formulas or ranges, references to wrong cells, rows left out of calculations. What’s more, 86% of spreadsheet creators are confident their files are correct, and individual inspections catch only 63% of errors. This means many strategic decisions are based on data that’s simply wrong.
But even accurate data won’t support decision-making if you can’t cross-reference it. When three departments record the same product name three different ways — one uses an internal code, another a trade name, a third an abbreviation — each is correct within its own convention. To the spreadsheet, though, these are three different products. Combining that data in a report ends in either manual line-by-line comparison or a flawed report that slips through unnoticed.
Excel stays consistent only when everyone sticks to the same rules, and with multiple sheets and a dozen people, that discipline breaks down sooner or later. Within a single sheet you can set up shared conventions — dropdown lists, fixed values — but even those usually lose to the realities of daily work.
On top of that, Excel conflates the data with how it looks on screen. A merged cell holding two types of data is unreadable to anything except a human eye. A pivot table won’t process it, and a formula can’t extract a single value from it. Excel works well for handling data or for displaying it — not both at the same time. With every formatting tweak, the sheet holds less data you can actually rely on.
Version chaos is a separate problem. Every copy of a spreadsheet sent by email, edited, and passed along is potentially a different version of the data. The more copies, the harder it is to identify the actual source of truth. Before making a decision, someone first has to figure out which version is current — and that means comparing files manually. There’s also no audit trail: Excel doesn’t record who changed what and when. If an error appears somewhere in the sheet, it’s hard to trace where it came from and how far it spread.
The more sheets, people, and processes, the more situations like these pile up. Each one alone may seem like a minor inconvenience, but together they mean skilled people spend their time verifying data and putting out fires rather than doing work that actually moves the organization forward.
What risks come with running your organization on Excel
Excel risks for business include losses that don’t show up in any spreadsheet: missed opportunities, confidential data leaks, and downtime when the one person who knows the sheet isn’t available.
You don’t know how much you’re losing — or how much you could gain
Your spreadsheets have been collecting years of data: connections between orders and inventory levels, patterns in purchase history, correlations between seasonality and raw material consumption. All of it could point to growth opportunities and risks. But without a tool that surfaces those patterns, you’re reacting rather than anticipating — seeing yesterday’s state, maybe today’s, but not the trend.
Take a Polish ultra-fresh food producer that planned production in scattered spreadsheets. When the data from Excel moved into a shared production planning system, one of the first orders triggered an alert: two tons of fresh spinach weren’t needed. The system monitored inventory in real time and caught what’s easy to miss at that order volume. The planning team stopped checking every order line by line and could focus on the work they were actually hired for.
The benefit didn’t stop at individual orders. The company also gained the ability to plan production 52 weeks ahead, so it could negotiate supplier terms in advance and match production capacity to seasonal demand shifts.
Your data isn’t as safe as you think
In 2023, the Police Service of Northern Ireland, responding to a routine staffing query, publicly released an Excel spreadsheet. A hidden tab in that file contained the personal data of 9,483 officers: names, ranks, and positions. Before anyone noticed, the file had been publicly accessible for over two hours — long enough for the data to reach paramilitary groups. The chief constable resigned, and the UK data protection regulator fined PSNI £750,000 — reduced from £5.6 million because it’s a public institution.
Similar incidents in everyday business aren’t hard to imagine. An Excel file, once sent by email, takes on a life of its own. You have no way to check who opened it, who forwarded it, or what they did with it. An employee might drop it into ChatGPT to speed up a report, and suddenly confidential data lands on servers you have zero control over. A security policy sets the rules, but it doesn’t enforce them on its own. To actually control how data moves through your organization, you also need tools that log access, restrict sharing, and minimize risk.
Your most critical processes depend on one person
In many organizations, there’s someone who knows “that Excel” inside out — they know which formulas depend on what, why column C is hidden, and where the numbers in the summary come from. It’s usually someone with deep knowledge of the business process who built the sheet in the first place.
Without them, nobody can fix a broken formula or explain why the report numbers look wrong. If they’re on vacation or sick leave and something breaks, the process waits at best. A tool meant for the whole team ends up depending on one person’s availability. The same applies to anything that requires interpretation. Routine entries? Anyone can handle those. But when an exception comes up — a non-standard order, a sudden change — Excel won’t tell you what to do. The sheet stores data, but the logic behind the organization’s operational decisions exists only in someone’s head.
And when that person leaves, the organization doesn’t have one problem — it has two: who replaces them, and how to keep the process running before anyone learns the sheet well enough to manage it.
Why organizations wait so long to make the switch
The decision to move processes out of Excel rarely happens proactively. It’s usually triggered by a breaking point — when the problems have piled up too far to ignore, or an incident that hits profitability or reputation.
That caution is understandable. If you’ve been through an implementation that took longer and cost more than planned, you have good reason to approach the next one carefully. But caution has a price too — and it isn’t implementation risk. It’s losing control over when and how the change eventually happens.
Waiting doesn’t remove the problem. Spreadsheet complexity and interdependencies grow, data and processes accumulate, and people settle deeper into their current way of working. If you still have room to choose the scope, pace, and timing of change, that’s a real advantage. In a few years, you may not. And when everything has to go in fast, projects become far more prone to errors, delays, and budget overruns.
How to find out which Excel processes cost you the most
To see which processes are running up the biggest tab, you need to do three things. First, inventory the spreadsheets that specific processes can’t run without. Second, assess each of those processes for operational complexity and business impact. Third, calculate how many hours per week your people spend manually maintaining those processes and what that costs. You’ll also need to factor in losses from spreadsheet errors.
Excel process inventory assistant
To help you with this, we’ve built an AI assistant that walks you through the inventory step by step. Based on your answers, it maps out what can stay in Excel, what’s worth putting on the change roadmap for the coming year, and which processes carry enough risk that it’s worth acting now. The assistant doesn’t require access to your systems or confidential data.
The assistant is a starting point that works for most organizations. If you want to go deeper into a specific process, reach out. A 30-minute conversation is where we start — enough to understand your situation and figure out what a detailed analysis would look like.
Sources:
Packy McCormick, Excel Never Dies, Not Boring, 2021
Raymond Panko, What We Know About Spreadsheet Errors, Frontiers of Computer Science, 2023
Steve Snyder (CEO, TransAlta), Human error costs TransAlta $24-million on contract bids, The Globe and Mail, 2003
Information Commissioner’s Office, What price privacy? Poor PSNI procedures culminate in £750k fine, ICO, 2024
