A spreadsheet is the most successful business tool ever made, and that is exactly why it becomes a liability. It asks for nothing. It starts empty and does whatever the person in front of it decides. For a young company that is a gift. The spreadsheet grows with the business, absorbs every new requirement, and never sends an invoice. For a while it is the reason the company can move at all.
The cost arrives later, and it arrives quietly, as time instead of a line item. Time is easy to overlook because everyone is busy. Someone copies numbers from one sheet into another every morning. Someone reconciles two versions that drifted apart. Someone rebuilds a report because a formula broke and no one is sure who last touched it. None of these tasks is large. Together they can consume a meaningful share of a capable person's week, and that share compounds as the business grows.
There is a second cost that is harder to see and more dangerous. A spreadsheet has no memory of who changed what, and no rule that stops a value from being wrong. A single mistyped figure propagates through every formula that references it, and the company makes decisions on the result without knowing the ground shifted. We have watched teams argue about which version of a file is current while the actual answer sat in a third copy nobody opened.
A spreadsheet presents the wrong number with exactly the same confidence as the right one.
The third cost is the one that eventually forces the conversation. Knowledge that lives in a spreadsheet lives in the person who built it. The formulas encode decisions that were never written down anywhere else. When that person is on holiday, the operation slows. When that person leaves, part of the company leaves with them. A business that runs on spreadsheets has quietly concentrated its most important logic in the least durable place.
None of this is an argument against spreadsheets. They are the right tool for thinking, for a quick model, for a question you will ask once. The failure is structural: a spreadsheet gets asked to serve as a system of record, a job it was never built for. It has no enforced structure, no reliable history, no rules that hold when a tired person is working fast at the end of a long day.
The honest way to measure the cost: count what the company gives up, because the license fee is nothing. Count the hours spent moving data by hand and keeping copies in agreement. Add the decisions made on figures that turned out to be wrong. Add the risk carried in the fact that one person understands the file everything depends on. For a growing company these add up to five figures a year without much effort, and the number rises faster than revenue does, because the manual work grows with volume while the business tries to grow with margin.
The point at which a spreadsheet should become a system arrives as a moment of recognition, at no particular size. It arrives when more than one person needs the same truth at the same time, when a wrong number would be expensive, or when the operation would stall if a particular file were lost. At that point the spreadsheet has finished its job well, and keeping it in that role stops being thrift. It becomes a bet that the manual work will not grow, placed by a company whose entire plan is to grow.
We replace a spreadsheet only when the operation has outgrown what it can safely hold, and we move only the parts that have earned a system. The rest can stay exactly where it is.