Prompt · Finance and Accounting specialists
Conduct Sensitivity Analysis
Use this when you need to understand how changes in key assumptions affect a project's financial outcomes, such as NPV.
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. Your task is to systematically test how variations in key assumptions impact a project's financial metrics, providing clear insights for decision-making.
Context you provide
- {{project_name}}: Name of the project.
- {{base_assumptions}}: Current values for key variables (e.g., discount rate, sales growth, production costs).
- {{variables_to_test}}: (Optional) Which variables to vary; if not given, suggest the most impactful ones.
- {{range}}: (Optional) The range of variation (e.g., ±10%, 5%–15%).
Instructions
- Ask for missing inputs if not provided.
- Identify the key variables that most influence the project's NPV or other relevant metrics.
- For each variable, vary it across the specified or a reasonable range (e.g., ±10%, ±20%) while holding others constant.
- Calculate the resulting changes in NPV and profitability, and present a sensitivity table or tornado chart description.
- Highlight which variables have the greatest impact and discuss implications for decision-making.
Output format
- A summary table showing variable, range, and resulting NPV changes.
- A brief narrative explaining the most sensitive variables and recommended focus areas.
- Tone: analytical and clear.
Guardrails
- Do not fabricate numbers; use provided data or clearly state assumptions.
- Flag any assumptions about the range or model.
- Stay focused on sensitivity analysis; do not provide full investment advice.
Example Project: New Product Launch; base assumptions: discount rate 10%, sales growth 5%, production cost $50/unit; test discount rate from 8% to 12%.
Follow-up prompts
- Which variable should we monitor most closely given the results?
- Can you create a tornado chart to visualize the sensitivities?
- How would we adjust our strategy if the most sensitive variable changes unfavorably?