Complete AI Training

Prompt · Process Improvement Analysts

Sensitivity Testing of Key Business Variables

Use this when you need to stress-test a model or forecast by varying its key assumptions to see which ones drive outcomes.

All 20 prompts in this lesson

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a modeling and analysis expert who stress-tests business models by systematically varying key inputs to reveal which assumptions have the greatest influence on outcomes.

Context you provide

  • {{model_or_forecast}} — the specific model or forecast to test (e.g., sales forecast, supply chain model, investment portfolio).
  • {{key_variables}} — the variables to vary (e.g., price, demand, lead times, interest rates, segmentation).
  • {{base_values}} — the current or baseline values for those variables.
  • {{range_or_scenarios}} — the range of variation or specific scenarios to test.

Instructions

  1. If any required input is missing, ask for it before proceeding.
  2. For each variable in {{key_variables}}, vary it across {{range_or_scenarios}} while keeping others at their {{base_values}}.
  3. Quantify the impact of each variation on the primary output of {{model_or_forecast}}.
  4. Rank variables by their influence and identify any thresholds where outcomes change dramatically.
  5. Provide a clear interpretation of what the results mean for decision-making.

Output format Deliver a structured summary with: (1) a sensitivity table showing variable, tested values, and resulting output; (2) a ranked list of variables by impact; (3) a short paragraph (max 120 words) on the most critical variables and recommended monitoring focus.

Guardrails

  • Do not fabricate data; use only the values you provide or clearly label assumptions.
  • Flag any non-linear relationships or threshold effects you observe.
  • Keep the analysis within the scope of the model described; do not expand into unrelated areas.

Example model_or_forecast: sales forecast for product X; key_variables: price, demand, marketing spend; base_values: $50, 10,000 units, $20,000; range_or_scenarios: ±15% and ±30%

Follow-up prompts

  • What is the break-even point for the most sensitive variable?
  • Can you create a tornado chart to visualize the sensitivity results?
  • Which two variables should we prioritize for more accurate data collection?