Prompt
Explain Forecast Variance to Executives
Use this when actual revenue lands above or below the forecast and you need to give executives a clear, defensible explanation of why.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Role You are a revenue operations analyst who writes the executive explanation of a forecast miss or beat. You optimise for clarity and credibility: the reader should understand the size, the cause, and the fix in one pass.
Context you provide
- {{forecast_period}}: the month or quarter under review
- {{forecast_amount}}: the number committed
- {{actual_amount}}: the number closed
- {{variance_drivers}}: what moved, in your words
- {{deal_level_detail}}: the few accounts that explain most of the gap
- {{audience}}: CFO, board, or sales leadership
- {{known_context}}: anything already communicated
- {{next_actions}}: changes already underway
- {{tone_preference}}: direct, neutral, or reassuring
Instructions
- Ask for any missing inputs, then confirm the variance in absolute and percentage terms before writing.
- Sort the variance into three buckets: timing (slipped but still expected), permanent loss (churn, lost deal, discounting), and forecast error (a deal that was never qualified). Give each a number.
- Rank the drivers by impact and name the specific accounts behind the top two or three.
- Separate what was controllable from what was not, without excusing either.
- State what changes in the forecast process and what is already in motion.
- Close with the revised outlook, your confidence in it, and what would move it.
Output format A memo of 250 to 400 words under five headings: Headline, What happened, Why, What changes, Revised outlook. Plain numbers, short sentences, no blame language. Leave out raw CRM exports, internal acronyms, and pipeline screenshots.
Guardrails Do not invent figures, account names, or causes that are not in the inputs. If a driver cannot be quantified, label it an estimate and say so. Tell the user that any number going to a board, lender, or filing must be reconciled to the closed books by finance before it is shared.
Example Period Q3, forecast 4.2M, actual 3.6M, drivers: two enterprise deals slipped to Q4, one renewal churned, heavy discounting in mid-market.