As of 13 August 2026, AI can create a budget versus actuals dashboard.

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, power-user skill, and roughly 1 hour until you can act on the result.

   How to actually do it

   1. Open the spreadsheet or accounting export containing the approved budget and actual transactions, and make sure each row has a date or period, category, amount and a budget or actual label.
   2. Create separate tabs or files for the budget and actuals, then remove accidental duplicate rows without changing legitimate repeated transactions.
   3. Paste both datasets into the prompt, including the reporting period, category rules and whether lower spending is favourable for each cost category.
   4. Ask the model to return the dashboard structure, formulas, variance tables, chart specifications and a separate list of data-quality problems rather than filling gaps.
   5. Put the returned tables and formulas into Excel or Google Sheets, keeping the original budget and actuals in separate source tabs.
   6. Compare the dashboard totals and category totals with the source reports, then test at least one favourable and one unfavourable variance by recalculating it manually.
   7. Check every flagged category and period with the person responsible for the accounts before using the commentary in a management meeting.
   8. Send the completed dashboard to the relevant manager or finance colleague with the source reports and a note listing unresolved data-quality issues.

   Prompt

   Create a budget versus actuals dashboard from the data below for a UK business. Use pounds sterling and do not invent, estimate or silently alter any figures. The budget covers [PERIOD], the actuals cover [PERIOD], and the reporting frequency is [MONTHLY OR OTHER]. The budget categories are [CATEGORY LIST OR USE THE CATEGORIES IN THE DATA].
   
   Budget data:
   [PASTE BUDGET DATA]
   
   Actuals data:
   [PASTE ACTUALS DATA]
   
   Build a spreadsheet-compatible dashboard specification with:
   1. Total budget, total actuals, absolute variance and variance percentage.
   2. The same measures by category and reporting period.
   3. Separate treatment for favourable and unfavourable variances, using the rule that lower spend is favourable for costs unless the data says otherwise.
   4. Charts for total budget versus actuals over time and the largest variances by category.
   5. A clear list of unmatched categories, missing periods, duplicate rows, blank values and other data-quality issues instead of guessing how to fix them.
   6. The formulas or calculation logic for every summary measure.
   7. A short plain-English commentary identifying only patterns directly supported by the data.
   
   Keep the source figures traceable to the dashboard. Show the assumptions separately, label all figures as budget or actual, and ask questions before calculating if the data structure is ambiguous. Do not provide accounting, tax or investment advice. This is not professional advice.

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