Guide · Spreadsheets, automation and AI
5 signs your business has outgrown spreadsheets, and where AI fits in
A practical guide with a worked example, a live dashboard and a free sample workbook you can download and test yourself.
The spreadsheet that quietly became your system
Almost every small business starts on a spreadsheet. It's free, flexible and everyone knows how to use it. That's the right choice at the start.
Then the business grows. The sheet that tracked ten orders now tracks a thousand. Three people edit it. Someone builds a formula nobody else understands. Without anyone deciding it, the spreadsheet has become your business system.
The good news: you now have more options than "keep struggling" or "buy expensive software". AI is built into Google Sheets and Excel, and simple automation tools can connect your apps. Used well, they can stretch a spreadsheet much further, or show you clearly when it's time to move on.
The 5 signs, and where AI helps
1. Nobody trusts the numbers
There are three versions of the same file, totals don't match, and meetings start with "which sheet is right?". Customer names are typed several ways, so reports split one customer into many.
Where AI helps: ask Gemini in Google Sheets or Copilot in Excel to find duplicates and standardise names. Then fix the cause: one master sheet, with dropdown lists so names can only be entered one way.
2. One person holds it all together
Only one person understands the formulas. When they're off sick or leave, nobody can produce the monthly numbers.
Where AI helps: both tools can explain a formula in plain English and suggest fixes for formula errors. Use that to write a one-page "how this sheet works" note, so the knowledge isn't stuck in one head.
3. Too much copy-pasting
The same order is typed into a sheet, an invoice, an email and a WhatsApp message. It wastes hours every week, and every copy is a chance for a mistake.
Where automation helps: tools like Zapier, Make or n8n move data between apps on their own. For example, a website form adds a row to your sheet and emails you an alert. We use exactly this pattern for our own enquiries.
4. People overwrite each other
Two people edit at once, a row disappears, and nobody knows who changed what, or when.
What to do: this is often the clearest sign you've outgrown the spreadsheet. A proper tool, such as a CRM, an accounting package or a simple database, gives each person the right access and keeps a history.
5. Simple questions take an afternoon
"Which customers haven't paid?" or "What sold best last month?" means an hour of filtering and copying.
Where AI helps: with clean data, you can ask questions in plain English and get a summary or chart. With messy data, AI just gives you a faster wrong answer, which is why signs 1 to 4 come first.
A worked example: Northgate Office Supplies
To show what this looks like in practice, I built a sample workbook for a fictional business. Every name and number in it is made up.
Before: a typical order sheet after a year of growth, with 24 orders. Dates are typed eight different ways. Kent Dental appears under four spellings. Quantities say "5 boxes" and "x2". Totals are typed by hand, and one order just says "see invoice". Someone added an "October total" row in the middle of the data, and a colleague keeps a second copy called "FINAL v2" that disagrees with the first. The total at the bottom says £6,553.00, and someone has written "(checked)" under it.
After: the same 24 orders, restructured. Customers, items and status come from dropdowns. Prices and due dates fill in automatically. The real total is £4,046.40. The messy sheet overstated sales by £2,506.60, and nothing on it looked wrong.