As of 13 August 2026, AI can analyse your historical business spending.

What to actually do

1. Hand it to a person

   The route this page recommends

   A person who owns the outcome does this end to end, worth it when the failure is dear.

2. Use a tool built for this

   Second choice

3. Do it yourself

   The distant third

   A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.

   How to actually do it

   1. Export the relevant business bank and card transactions for the period from your banking or accounting system as CSV or XLSX, including transaction date, description, amount, currency and account.
   2. Open the export in a spreadsheet and remove personal transactions, blank rows and obvious duplicates without changing the original amounts or descriptions.
   3. Add a notes or category column only where you already know the business purpose, and keep unclear transactions marked as unclear rather than guessing.
   4. Upload the cleaned spreadsheet to an AI data analysis tool and paste the prompt, including the business type, accounting basis and reporting period.
   5. Ask the tool to produce transaction-level checks, supplier and category summaries, monthly totals, unusual movements and a list of uncertain items.
   6. Recalculate the total and monthly subtotals in your spreadsheet and compare them with the AI output, then match a sample of supplier totals back to the original bank or card statements.
   7. Send the flagged transactions and the final spending summary to your accountant before using uncertain categories for VAT, tax records, management accounts or a formal forecast.

   Prompt

   Analyse the attached historical business spending data for a UK business. Use only the supplied data and do not invent transactions, categories, VAT treatment, suppliers or explanations. First report the date range, number of transactions, total spend and any missing or duplicated rows. Then group spending by supplier and by category, showing totals and percentages, and identify monthly trends, unusual changes and recurring costs. Preserve the original amounts and currency. Flag transactions whose category, business purpose or VAT treatment is uncertain instead of guessing. Separate factual findings from possible explanations. Reconcile every summary total to the transaction-level data and show the formulas or calculation method used. Finish with a short list of questions I should answer before using this analysis for budgeting or management accounts. This is management information, not professional advice, so do not decide whether any item is tax-deductible or VAT-reclaimable. Business context: [business type]. Accounting basis: [cash or accrual, if known]. Existing categories: [paste categories, if any]. Reporting period: [start date to end date].

   Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.