Seven Signs Your Business Has Outgrown Its Spreadsheets

Nobody decides to run their business on a spreadsheet. It happens gradually: a file created to answer one question turns out to be useful, someone else starts using it, and a few years later it is holding something the company depends on.

The difficulty is noticing. There is no moment where it stops working. It just gets slower, more fragile, and more dependent on one person. These are the signs that the line has been crossed, and what each one actually costs.

1. The file has a version in its name

Pipeline_v4_FINAL_updated_JB.xlsx is not a naming convention, it is a symptom. It means there is no single authoritative copy and people have resorted to labelling their way out of the confusion.

What it costs: time spent working out which file is current, and decisions made from a version that was not. The expensive version of this is two people acting on different numbers and neither realising until later.

What to look for: count how many copies exist across email, shared drives and laptops. The number is usually higher than anyone expects.

2. Someone reconciles it against another system

If part of a routine involves exporting from one place and pasting into another, or checking that two systems agree, the spreadsheet is acting as an integration layer held together by a person.

What it costs: the hours directly, plus the errors that reconciliation exists to catch. And it never ends. The task recurs weekly or monthly, forever, and it grows with the business.

What to look for: ask whose calendar has a recurring block for it.

3. Only one person really understands it

Most long-lived operational spreadsheets have an author. They built the formulas, they know which tab feeds which, and they know the two columns that must not be sorted. When they are on leave, changes wait.

What it costs: key-person risk, in its purest form. If they leave, the business keeps running a system nobody can safely modify. This is the single most common reason a spreadsheet finally gets replaced, and waiting for that moment is the most expensive way to do it.

What to look for: ask who would make a change if the usual person were away for a month. If there is no confident answer, that is the finding.

4. It contains data that should be restricted

Salaries, client contact details, pricing, anything personal. A spreadsheet has one access level: whoever can open the file can read all of it. Hidden columns and sheet protection are conventions, not controls.

What it costs: under UK GDPR you are expected to limit access to personal data to those who need it, and to be able to say who has it. A shared workbook makes that question hard to answer honestly, and harder still to evidence.

What to look for: open the sharing settings and read the list of people with access. Then ask how many of them need every column.

5. Nobody can say what changed, or who changed it

When a number is wrong, the useful questions are what it was before, when it changed, and on whose authority. File-level version history gives a partial answer. It will not tell you that a discount on one line was altered on a Tuesday and by whom.

What it costs: investigations that take hours instead of seconds, and disputes that cannot be settled with evidence. In a regulated context it is worse than inconvenient.

What to look for: pick a figure that changed recently and try to establish its history. Time how long it takes.

6. The process depends on someone remembering

A spreadsheet does not chase. It does not notice that an approval has been sitting for a week, that a renewal falls due next month, or that a record was started and never finished. Any of that has to live in a person’s head or a reminder they set.

What it costs: things slip. Renewals lapse, approvals stall, half-finished records sit in the file looking complete. The cost is invisible because the misses are never counted.

What to look for: filter for records that have not changed in ninety days. Ask whether they should have.

7. The monthly report is a job rather than a click

If producing the regular pack means assembling several files, fixing formatting, and checking the totals, the data is not in a usable shape. It is being made usable by hand, every time.

What it costs: a day or more a month of someone capable, plus the lag. By the time the numbers are ready they describe a situation that has moved on.

What to look for: ask how many hours the last one took, and how many people were involved.

Scoring it honestly

One or two of these is normal and not worth acting on. Spreadsheets are useful tools and a bit of friction is the price of their flexibility.

Three or four, and you are carrying a real cost that nobody has added up.

Five or more, and the spreadsheet is an operational system without any of the properties an operational system needs, and the question is no longer whether to replace it, but which part to replace first.

Working out what it is worth

Before anyone talks about software, put a number on the current position. For each sign you recognised, estimate the hours per month and multiply by a rough hourly cost. Add anything you can attribute to errors found late.

This figure is the whole business case. It is also, in our experience, the part people most often skip, and without it any quote is impossible to evaluate, because there is nothing to compare it against. We go through how to do that comparison in how much custom software costs in the UK.

What to do next

Do not replace everything. Take the one spreadsheet that would hurt most if it were wrong tomorrow.

Document what it does before changing anything. A spreadsheet that has run a process for years encodes every exception the business actually has, usually more accurately than anyone can describe from memory. That document is most of a specification.

Decide whether a product already does it. Sometimes one does, and that is the cheaper answer. Custom versus off-the-shelf sets out how to tell.

Keep the spreadsheet for what it is good at. Modelling, one-off analysis, thinking something through. It was always excellent at that. The problem was only ever asking it to run the business as well, which is covered in why spreadsheets slow a business down.