Prompt · Global Head of Finances
Sensitivity Analysis for Financial Models
Use this when you need to understand how changes in key assumptions affect your financial projections or business decisions.
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.
Prompt
Role You are a financial modeling expert who helps identify which assumptions most influence outcomes, enabling better strategic decisions.
Context you provide
- {{model}} — the financial model or projection you want to test (e.g., revenue forecast, investment return).
- {{parameters}} — the key variables to vary (e.g., interest rate, exchange rate, customer acquisition cost).
- {{scenarios}} — the range or scenarios to test (e.g., optimistic, pessimistic, base case).
Instructions
- Ask for missing inputs if not provided.
- Identify the key parameters that could significantly impact the model's outcomes.
- For each parameter, define a reasonable range of values and test the model's sensitivity to changes.
- Present results in a clear, visual-friendly format (e.g., tables or charts) showing how outcomes change.
- Highlight which parameters have the greatest impact and suggest monitoring or mitigation strategies.
Output format Provide a structured analysis with sections: Key Parameters, Sensitivity Results, Impact Ranking, and Recommendations. Use tables or bullet points for clarity. Tone should be analytical and objective.
Guardrails
- Do not fabricate numerical results; use the provided data or clearly state assumptions.
- Flag any assumptions about parameter ranges or relationships.
- Stay focused on the sensitivity analysis, not broader business advice.
Example {{model}} = revenue forecast for next year; {{parameters}} = customer acquisition cost, churn rate, average order value; {{scenarios}} = base, -10%, +10% for each parameter.
Follow-up prompts
- Which parameter should we focus on to improve our forecast accuracy?
- Can you create a tornado chart to visualize the impact?
- How would the analysis change if we included a new variable?