As of 13 August 2026, AI can only partly forecast revenue from your sales leads.

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 your CRM or sales spreadsheet and export the current opportunities with opportunity ID, value, currency, stage, probability, expected close date, sales owner and last activity date.
   2. Remove duplicate records and check that each opportunity has a current value, a close date and a clearly defined stage before pasting the export into the prompt.
   3. Gather comparable historical actual revenue and your written definitions of committed, best-case and uncertain opportunities, then add them to the relevant prompt sections.
   4. Paste the completed prompt and data into a chatbot, and ask it to produce the forecast without adding assumptions that are not in your data.
   5. Copy the output into a spreadsheet and recalculate the expected revenue totals, monthly groupings and scenario totals from the opportunity rows.
   6. Ask your sales manager or sales operations colleague to compare the classifications, probabilities, close dates and largest opportunities against the CRM and recent customer conversations.
   7. Record the approved assumptions and send the checked forecast to the people responsible for targets, hiring or spending, clearly labelled as a forecast rather than booked revenue.

   Prompt

   Act as a cautious sales operations analyst. Forecast revenue from the lead and opportunity data I paste below.
   
   Use only the data supplied here. Do not invent conversion rates, deal values, close dates, sales cycles or customer information. First identify missing, inconsistent or stale fields. Then:
   1. Calculate the expected revenue using the supplied probability for each opportunity, showing the formula and totals.
   2. Produce a conservative, central and optimistic scenario. If the data does not contain defensible assumptions for these scenarios, state that clearly and show which assumptions would be needed rather than making them up.
   3. Group the result by expected close month and sales owner, where those fields exist.
   4. Separate committed, best-case and uncertain opportunities using the definitions in the data. If no definitions are supplied, ask me for them before classifying.
   5. List the opportunities that have the greatest effect on the total and explain why they make the forecast fragile.
   6. Compare the result with any historical actual revenue I provide, but do not claim predictive accuracy from a single comparison.
   7. End with a short list of checks a sales manager must complete before using this forecast for targets, hiring or spending.
   
   Return tables that can be copied into a spreadsheet. Keep currency in pounds and state whether figures are gross or net of VAT. Treat this as decision support, not a confirmed revenue figure.
   
   Business context:
   [Sales period, reporting currency, VAT treatment, and definitions of committed, best-case and uncertain]
   
   Lead or opportunity data:
   [Paste a CSV or table with opportunity ID, value, currency, stage, probability, expected close date, sales owner, last activity date and any other relevant fields]
   
   Historical actual revenue, if available:
   [Paste comparable historical periods and explain how they were measured]

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