April 22th, 2026
What Is a Sensitivity Analysis?: Complete Guide with Examples
By Simon Avila · 15 min read
Understanding what a sensitivity analysis is can change how your team approaches forecasting, budgeting, and risk. I tested it across financial models and business planning scenarios to give you a full breakdown of how it works and how to run one.
What Is Sensitivity Analysis?
A sensitivity analysis is a method for testing how changes in one variable affect the outcome of a model or forecast. You pick a key input, like revenue growth or ad spend, change it by a set amount, and see what happens to your output. It's often called "what-if" analysis for that reason, and it's one of the more practical ways to stress-test a business decision before you commit to it.
Types of sensitivity analysis
Not every sensitivity analysis works the same way. The method you use depends on how many variables you're testing and how much complexity your model can handle. Here are some of the most common types you'll come across:
One-at-a-time (OAT) analysis: One-at-a-time (OAT) analysis is the simplest approach. You change one variable at a time while keeping everything else fixed, then record the result. It's a good starting point because it requires no specialist software and gives you a clear, readable output for each variable, though it won't capture how variables interact with each other.
Multi-variable analysis: Instead of testing one input at a time, you change several variables at once. This can give you a more realistic picture of how your model behaves, since real-world conditions rarely change in isolation. I find this approach works better once you've already identified your key variables through a simpler OAT analysis first.
Scenario analysis: You define a specific set of conditions, like a best case, worst case, and base case, and run your model against each one. This is closely related to sensitivity analysis but focuses on predefined situations rather than individual variables.
Monte Carlo simulation: Monte Carlo simulation runs your model hundreds or thousands of times using randomly generated input values, producing a range of possible outcomes with probabilities attached. It's more complex to set up than a basic OAT analysis, so I find it's best saved for high-stakes decisions rather than routine forecasting.
Tornado diagrams: A visual method that ranks your variables by impact. The inputs with the biggest effect on your output appear at the top, making it easier to see where your model is most exposed.
How to do a sensitivity analysis
Sensitivity analysis works by changing one input variable at a time and tracking how your output responds. You don't need specialist software to get started, a spreadsheet can handle most basic analyses. Here's the basic process:
Set your baseline: Start with your current model and its assumptions, like your projected revenue, costs, or conversion rate. This is your control, the number against which everything else gets measured.
Pick a variable to test: Choose one input to adjust, like price, demand, or ad spend. I recommend starting with the variables you're least certain about.
Change it by a set amount: Move it up or down by a realistic percentage. A common starting point is 10% in each direction, but you can adjust this based on how volatile that variable tends to be.
Record the output: Note how your result changes each time. A big swing in output means that the variable carries more risk in your model.
Repeat for each variable: Work through your key inputs one at a time, keeping everything else fixed while you test each one.
By the end, you can see which variables shift your outcome the most, and which ones barely move it.
💡 Tip: If you're working with a more complex dataset, a tool like Julius can handle the analysis from a simple text prompt. That way, you can spend more time interpreting results than running them.
Examples of sensitivity analysis
Financial planning
Finance teams can use sensitivity analysis to stress-test their forecasts before presenting them to leadership. I've seen this come up most often around revenue projections, where a team wants to know what happens to their bottom line if sales come in 10% or 20% below target. It can also help model how changes in interest rates or operating costs might affect cash flow over time.
Marketing budgets
Marketers can use it to find out which parts of a campaign have the biggest impact on ROI (return on investment). For example, you might test how a 15% drop in conversion rate affects your cost per acquisition, or whether increasing ad spend by 20% is likely to produce a proportional return.
In my experience, this kind of analysis can help teams avoid spending their budget on campaigns built around untested assumptions.
Pricing decisions
If you're weighing a price change, sensitivity analysis can show you how much your revenue might shift depending on how customers respond. A 10% price increase might look attractive on paper, but if customers are sensitive to price changes, the actual impact on revenue could go either way.
Operational planning
Operations teams can model how changes in supply costs, lead times, or production volume might affect margins. This can be useful when evaluating new suppliers or planning for potential disruptions in your supply chain.
Investment analysis
Analysts can use sensitivity analysis to test how changes in key assumptions might affect the projected value of an investment. For example, a private equity team evaluating an acquisition might test how the deal looks if revenue grows at 5% instead of 10%, or if margins compress by a few points.
Rather than presenting a single number, you can show a range of outcomes depending on which assumptions hold.
Product decisions
5 best tools for sensitivity analysis
Finding the right tool for sensitivity analysis depends on your technical comfort level and how complex your models are. I tested several options across different use cases. Here are the 5 best for business teams:
Julius: An AI-powered data analysis tool that can run sensitivity analyses from a natural language prompt, no coding required. You can upload a dataset, connect a data source, or pull financial data directly in the platform, then ask a question in plain English and get a chart or graph of your results. It can be a good fit for teams that want to run analyses without relying on a data analyst.
@RISK by Lumivero: A dedicated risk and sensitivity analysis tool that runs Monte Carlo simulations directly inside Excel. It can generate tornado diagrams and spider charts to visualize which variables carry the most risk, though it may be more than most business teams need if you're just getting started.
Microsoft Excel: The most widely used tool for basic sensitivity analysis, with built-in What-If Analysis, Data Tables, and Goal Seek functions. It works well for straightforward one-at-a-time analyses, though it can get unwieldy as your model grows in complexity. The learning curve can be steep, and it works best when paired with a tool that handles the underlying analysis.
Power BI: A business intelligence platform that can help visualize sensitivity analysis results through interactive dashboards and reports. It works well when you need to share findings with stakeholders who want to explore the data themselves.
Python: A flexible programming language with libraries like pandas and SALib that can handle more advanced sensitivity analysis for technical users. It takes more setup than a spreadsheet, but offers a lot more control over how your analysis is structured and visualized.
Benefits of sensitivity analysis
Sensitivity analysis won't make decisions for you, but it can make the decisions you do make a lot easier to defend. Here are some of the main benefits:
Identifies your riskiest assumptions: By testing variables one at a time, you can spot which inputs affect your outcome the most. In my experience, this is often where teams find the assumptions they've been taking for granted.
Reduces uncertainty: You can't eliminate uncertainty from a forecast, but sensitivity analysis can help you understand its shape, showing you the range of outcomes you might be working with.
Supports better decision-making: When you can see how a model responds to different inputs, it's easier to make a call with confidence, or at least understand the trade-offs involved. I find it useful when a decision is close, and you need something more concrete than gut feeling.
Makes forecasts easier to communicate: A range of outcomes is often more useful to stakeholders than a single number. Sensitivity analysis gives you the data to show best-case, worst-case, and middle-ground scenarios side by side.
Helps prioritize resources: Once you know which variables matter most, you can focus time and budget on managing those inputs rather than spreading attention evenly across everything. I've seen this save teams a significant amount of time during planning cycles.
Want to run a sensitivity analysis without the hassle? Try Julius
Knowing what a sensitivity analysis is and being able to run one quickly are two different things. Julius is an AI-powered data analysis tool built to make that second part easier. You can pull financial data directly into the platform, ask "what if" questions in plain English, and get clear visual outputs without any coding required.
Here’s how Julius helps:
Data search: Type your question, and Julius can search for relevant public data or pull live financial market data for over 17,000 companies through its Financial Datasets integration, so you don’t need a file or database connection to begin.
Direct connections: Link databases like PostgreSQL, Snowflake, and BigQuery, or integrate with Google Ads and other business tools. You can also upload CSV or Excel files. Your analysis can reflect live data, so you’re less likely to rely on outdated spreadsheets.
Repeatable Notebooks: Save an analysis as a notebook and run it again with fresh data whenever you need. You can also schedule notebooks to send updated results to email or Slack.
Smarter over time: Julius includes a Learning Sub Agent, an AI that adapts to your database structure over time. It learns table relationships and column meanings as you work with your data, which can help improve result accuracy.
Quick single-metric checks: Ask for an average, spread, or distribution, and Julius shows you the numbers with an easy-to-read chart.
Built-in visualization: Get histograms, box plots, and bar charts on the spot instead of jumping into another tool to build them.
One-click sharing: Turn an analysis into a PDF report you can share without extra formatting.
Ready to see how Julius can simplify sensitivity analyses? Try Julius for free today.
Frequently asked questions
What is the difference between sensitivity analysis and scenario analysis?
Sensitivity analysis tests how changes in one variable at a time affect your outcome, while scenario analysis looks at a specific set of conditions all at once. For example, a sensitivity analysis might ask "what happens if sales drop by 10%?" while a scenario analysis would model a full recession with multiple variables changing simultaneously.
What are the limitations of sensitivity analysis?
The main limitation of sensitivity analysis is that it tests variables one at a time, which means it can miss how variables interact with each other in the real world. It also relies on the assumptions built into your model, so if those assumptions are off, your results will be too.
Is sensitivity analysis the same as a what-if analysis?
Yes, sensitivity analysis and what-if analysis refer to the same thing. Both describe the process of changing input variables in a model to see how the output responds. The terms are used interchangeably across finance, business planning, and data analysis.
