When to replace a spreadsheet with real software

Almost every internal system we have built replaced a spreadsheet, and in most cases the spreadsheet had been doing the job for years. That is not a failure. It is a spreadsheet being used past the point it was designed for.

The question is not whether spreadsheets are bad. It is whether yours has quietly become a piece of software that nobody maintains. Checked 12 August 2026.

Eight signs it has outgrown the job

None of these on its own is decisive. Three or more together usually is:

• There are versions. stock_final_v3_USE_THIS.xlsx. The moment a filename carries version information, you have a database with no version control.
• Two people cannot safely edit it at once. Or they can, and you have silently been losing each other's changes.
• The same customer appears three times, spelled differently. Data that should be relational is being retyped instead of referenced.
• A formula breaks when someone inserts a row. VLOOKUPs pointing at shifting ranges are the classic symptom.
• Only one person really understands it. This is the dangerous one. The business logic lives in someone's head and in a tab nobody else opens, and it leaves when they do.
• Producing the monthly report takes hours. Copying between tabs to answer a question you ask every month is a job a machine should have.
• It holds personal data. A file on a laptop with customer details is a data protection problem waiting for an audit.
• Decisions that matter are made from it. Pricing, stock, payroll, compliance. The cost of it being quietly wrong is no longer small.

The costs that never reach an invoice

A spreadsheet looks free, which is why the comparison is usually unfair. What it actually costs:

• Staff hours spent typing things that could arrive already typed
• Decisions made from numbers that turned out to be a fortnight old
• Rework when two versions have to be reconciled by hand
• The risk sitting in one person's knowledge
• Compliance exposure on personal data nobody is tracking

Add up the hours honestly before comparing them to a build price. Two people spending an afternoon a week on manual entry is roughly 200 hours a year.

A worked example

An Italian estate agency runs a street-by-street property census, going door to door to find owners who might be ready to sell. That ran on five spreadsheets: properties, owners, visits, outcomes and a working copy somebody kept locally.

We replaced it with FileCasa, an internal web app. Same information, one place, no versions, and questions that used to mean an evening of copying now take a filter. That is the shape of this kind of project: not new capability, but the same work with the friction taken out.

When to keep the spreadsheet

We turn this work down fairly often. Keep it when:

• One person uses it and it works. Single-user spreadsheets that do their job are fine. Leave them alone.
• The process is still changing weekly. Software encodes decisions. If you are still working out what the process is, a spreadsheet's flexibility is a feature, and building now means building the wrong thing.
• It is genuinely a calculation, not a process. Modelling, forecasting, one-off analysis. That is what spreadsheets are for and nothing replaces them.
• An off-the-shelf tool already covers it. If it is standard bookkeeping or standard CRM, buy the thing. Custom is for the parts of your business that are actually specific to you.

What it costs and how long it takes

An internal web app is our largest shape: from £4,000, with most landing between there and £8,000, and a large platform going beyond. Bigger builds run several months rather than several weeks.

The usual approach is to start with the part that hurts most rather than replacing all five spreadsheets at once. A first version that removes the worst of it in six weeks beats a complete system in nine months, and it means you are using something real before the second half is designed.

Detail on prices is in the cost guide, and on timelines in the timeline guide.

Describe your spreadsheet