As of 13 August 2026, AI can only partly create a rolling business forecast.

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 1 hour until you can act on the result.

   How to actually do it

   1. Open your accounting system and bank records, then export the historical sales, receipts, costs, payments, payroll, VAT, tax, debt and capital-spending data used to run the business.
   2. Gather current commitments from supplier agreements, payroll records, loan statements, rent records, subscriptions and tax payment schedules, keeping their payment dates.
   3. Write down the assumptions you actually want to use for sales, prices, payment delays, supplier costs, wages, one-off items and planned spending, marking uncertain assumptions separately.
   4. Paste the cleaned tables, commitments and assumptions into the prompt, and ask the chatbot to produce the forecast with its formulas, scenarios and missing-input list.
   5. Open the generated spreadsheet or table and compare every historical figure with the accounting export and bank records, correcting any changed dates, duplicated transactions or omitted items.
   6. Recalculate the forecast using your own base, downside and upside assumptions, then compare the closing cash and major movements with what you know about upcoming trading and commitments.
   7. Ask a colleague or accountant to check the treatment of VAT, payroll, tax, debt, capital spending and unusual transactions before using the forecast for a borrowing, hiring or major spending decision.
   8. Save the approved version with its source data and assumptions, then update the actual period and roll the forecast forward whenever new results are available.

   Prompt

   Create a rolling business forecast from the data and assumptions below.
   
   Business: [business type]
   Currency: GBP
   Forecast period and update frequency: [for example, monthly forecast updated each month]
   Historical period covered: [period]
   
   Historical actuals:
   [paste a table with dates, sales by category, direct costs, overheads, payroll, VAT, tax, debt repayments, capital spending and opening cash where available]
   
   Known future commitments:
   [paste rent, salaries, supplier payments, loan payments, subscriptions, tax and other committed amounts with dates]
   
   Assumptions:
   [paste expected sales volumes, prices, payment timing, supplier costs, wage changes, one-off items and any other assumptions]
   
   Build a clear forecast using only the supplied data. Do not invent missing figures. Mark every estimate, assumption and missing input. Separate cash receipts and payments from profit-and-loss items, and explain the treatment of VAT, payroll, tax, debt and capital spending where relevant. Show the opening balance, closing cash, sales, direct costs, overheads and operating result for each forecast period. Include a base case, a downside case and an upside case, changing only assumptions that are explicitly identified. Add a sensitivity table showing which assumptions have the largest effect on closing cash. Preserve the source figures, show the formulas or calculation logic, and list the checks I must perform before relying on the forecast. State clearly where an accountant should review the model. Use UK English and pounds sterling.

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