1. Sensitivity analysis
Now, we'll take a deep dive into sensitivity analysis.
2. What is sensitivity analysis?
The most common approach in sensitivity analysis is to examine how changes in one variable at a time affect the final result while keeping all other factors constant. This is repeated for all variables under consideration. It helps identify the most influential drivers in a model, enabling businesses to focus on key factors that affect their performance.
Sensitivity analysis helps businesses prioritize key risks and opportunities by identifying which factors have the greatest impact on performance. By incorporating this technique into decision-making, organizations can pro-actively adjust strategies and enhance resilience against uncertainty.
3. Sensitivity analysis vs. scenario analysis
Sensitivity analysis might seem similar to scenario analysis, but they serve different purposes and are used in different ways.
Scenario analysis typically analyzes the combination of variables at the same time. On the other hand, sensitivity analysis focuses on how changes in specific variables affect the overall outcome.
While Scenario analysis helps businesses prepare for multiple possible futures, sensitivity analysis helps identify which variables have the biggest impact on results.
Scenario analysis will typically focus on evaluating best-case, baseline, and worst-case scenarios while sensitivity analysis provides a range of possible outcomes due to changes in one variable.
They are complementary however, and can be used together.
4. Steps of sensitivity analysis
Sensitivity analysis, like the other techniques in this chapter, follows a systematic approach.
First, you need to identify the key input variables you'll consider for the analysis.
5. Steps of sensitivity analysis
Next, you'll need to define a reference scenario, establish a reference point with typical values for all variables.
6. Steps of sensitivity analysis
The third step is to adjust one variable at a time, increase or decrease a single factor while keeping others constant to measure its impact.
7. Steps of sensitivity analysis
Finally, analyze the results and observe how the outcome shifts due to the change in input to determine which variables cause the most significant fluctuations and prioritize them for decision-making.
Should you combine scenario analysis with sensitivity analysis, these steps will overlap.
8. Visualizing the results of sensitivity analysis
The results of sensitivity analysis can be easily visualized. Common visualizations are tornado diagrams, spider charts, and heatmaps.
Tornado diagrams work best when comparing the impact of variables, either positive or negative. They are particularly useful for quickly assessing key risk factors.
9. Visualizing the results of sensitivity analysis
Spider charts are ideal for showing the relationships between multiple variables. They help in situations where trade-offs need to be considered, such as balancing cost, quality, and time in project management.
10. Visualizing the results of sensitivity analysis
Heatmaps work well when analyzing large datasets with multiple input variables. They are particularly useful for visually spotting patterns and correlations between variables, making them a go-to for analysts handling complex models.
11. Example: Profit margins
Say, a manufacturing company wants to understand how different cost and revenue factors impact its profit margins. To do this, they conduct a sensitivity analysis, testing how changes in product price, raw material cost, labor cost, and advertising spend affect overall profitability.
12. Example: Profit margins
We can see the results of that analysis in this tornado diagram. The longer the bar, the larger the influence of that variable. The bar to the left is the negative impact, the bar to the right the positive impact.
Selling price has the greatest impact. If the company wishes to boost profits, adjusting the selling price or lowering raw material costs would be the most effective strategy.
13. Let's practice!
Over to you!