A waterfall plot is one of those visualizations that looks simple at first glance but can reveal a story that a table of numbers often hides. It shows how an initial value increases or decreases through a sequence of positive and negative changes until it reaches a final value. Whether you are explaining company profit, customer churn, project costs, or model predictions, a waterfall plot helps answer one key question: What exactly caused the change?
TLDR: A waterfall plot visualizes how a starting number moves step by step to an ending number through gains and losses. For example, a business might start with $500,000 in revenue, add $120,000 from new customers, lose $45,000 from churn, and end at $575,000. In analytics, waterfall plots are especially useful because they turn percentage changes, budget shifts, or performance drivers into a clear visual story. They are commonly used in finance, marketing, sales, operations, and data science.
What Is a Waterfall Plot?
A waterfall plot, also called a waterfall chart or bridge chart, is a data visualization that displays cumulative changes between a starting point and an ending point. Each bar represents a contribution to the total. Positive values typically rise upward, while negative values fall downward.
The name “waterfall” comes from the way the bars appear to cascade across the chart. Instead of showing unrelated columns side by side, a waterfall plot connects each step to the one before it. This makes it ideal for showing cause and effect, not just comparison.
In a typical waterfall plot, you will see:
- Starting value: The baseline or opening amount.
- Intermediate changes: Positive and negative adjustments that move the value up or down.
- Ending value: The final result after all changes are applied.
- Color coding: Often green for increases, red for decreases, and blue or gray for totals.
Why Waterfall Plots Are Useful
The biggest advantage of a waterfall plot is clarity. Many reports show only the starting number and the ending number, leaving people to guess what happened in between. A waterfall plot fills that gap.
Imagine a company’s quarterly profit dropped from $250,000 to $210,000. A basic chart might show the decline, but it would not explain it. A waterfall plot could reveal that sales increased by $40,000, costs rose by $65,000, refunds reduced income by $10,000, and tax savings added back $5,000. Suddenly, the final number makes sense.
Waterfall plots are useful because they:
- Show how a total changed, not just that it changed.
- Make positive and negative effects easy to compare.
- Help identify the biggest drivers behind a result.
- Support better decisions in meetings, reports, and dashboards.
- Turn complex financial or operational data into a readable story.
Common Examples of Waterfall Plots
1. Financial Performance
Finance teams often use waterfall plots to explain revenue, profit, cash flow, or budget variance. For example, a company may begin the month with $1 million in projected revenue and then show how discounts, new deals, cancellations, currency effects, and refunds changed the final result.
This is especially useful in board reports because executives can quickly see which factors helped and which hurt performance. Instead of reading ten rows of numbers, they can visually understand the full financial bridge.
2. Marketing Campaign Analysis
Marketers can use waterfall plots to examine campaign performance. Suppose a campaign starts with 100,000 website visitors. Organic search adds 35,000, paid ads add 22,000, email contributes 9,000, but bounce rate removes 28% of potential conversions. A waterfall plot can show how traffic sources and losses lead to the final number of qualified leads.
This makes it easier to see whether a campaign’s growth came from one major channel or from several smaller improvements.
3. Sales Pipeline Changes
Sales teams often track how pipeline value changes over time. A waterfall plot can show the opening pipeline, newly created opportunities, upsells, lost deals, delayed deals, and closed revenue. This helps managers understand whether a weak sales month happened because of poor lead generation, low close rates, or deal postponements.
Image not found in postmeta
4. Customer Churn and Retention
Subscription businesses can use waterfall plots to explain changes in monthly recurring revenue. A company might start with $80,000 in recurring revenue, gain $12,000 from new subscriptions, add $6,000 from upgrades, lose $7,500 from cancellations, and end at $90,500. The chart makes growth and churn visible in one place.
5. Data Science and Machine Learning
In data science, waterfall plots are often used to explain model predictions. For example, in a credit risk model, the plot might show how income, loan amount, payment history, and credit utilization push a prediction higher or lower. This is common in model explainability, especially when using methods such as SHAP values.
Here, the “starting value” might be the average prediction, while each feature moves the prediction toward the final output. This helps nontechnical stakeholders understand why a model made a particular decision.
How to Create a Waterfall Plot
Creating a waterfall plot is not difficult, but it does require clean, well-structured data. The most important step is deciding the logical order of changes. A waterfall plot should read like a story from left to right.
Step 1: Define the Starting and Ending Values
Begin with a baseline. This could be starting revenue, opening balance, initial customer count, projected budget, or average model prediction. Then decide what final number the chart should reach.
For example:
- Starting revenue: $500,000
- New sales: +$95,000
- Upsells: +$35,000
- Cancellations: -$40,000
- Discounts: -$15,000
- Ending revenue: $575,000
Step 2: List Each Positive and Negative Change
Break the movement into meaningful components. Avoid using too many tiny categories, as this can make the chart crowded. If you have dozens of minor changes, group them under a label such as Other adjustments.
Step 3: Calculate the Running Total
A waterfall plot depends on cumulative movement. Each bar begins where the previous bar ended. If you start at 500,000 and add 95,000, the next value becomes 595,000. If you then subtract 40,000, the running total becomes 555,000.
This running total is what gives the plot its step-like appearance.
Step 4: Choose Colors Carefully
Color is not just decoration in a waterfall plot; it carries meaning. A common convention is:
- Green: Positive changes
- Red: Negative changes
- Blue, black, or gray: Starting and ending totals
Use consistent colors across reports so readers do not have to relearn the visual language each time.
Step 5: Add Clear Labels
Labels are crucial. Include category names, values, and possibly the final total. If your audience is not familiar with the data, add a short subtitle explaining what the chart measures and over what time period.
Image not found in postmeta
Best Practices for Waterfall Plots
To make your waterfall plot more effective, keep these tips in mind:
- Keep the sequence logical: Arrange changes in the order they occurred or in the order your audience expects.
- Do not overcrowd the chart: Too many bars can make the visualization hard to read.
- Use descriptive category names: Labels like “Cost increase” are clearer than vague terms like “Adjustment 2.”
- Highlight the largest drivers: Make it easy to spot the most important gains and losses.
- Check the math: The ending value must equal the starting value plus all intermediate changes.
When Not to Use a Waterfall Plot
A waterfall plot is powerful, but it is not always the right choice. If you simply want to compare unrelated categories, a bar chart may be better. If you want to show trends over many time periods, a line chart may be clearer. If you need to show proportions of a whole, a stacked bar chart or pie chart might work better.
Use a waterfall plot when your main goal is to explain a transition: how one value became another through a series of additions and subtractions.
Final Thoughts
A waterfall plot is a practical and engaging way to explain change. It turns a final number into a visual journey, showing every major increase and decrease along the way. From financial reports to marketing dashboards and machine learning explanations, it helps people understand not only the result, but the reasons behind it.
If your data has a clear starting point, a set of meaningful changes, and a final outcome, a waterfall plot may be the perfect visualization. It gives your audience something better than numbers alone: a clear story of what happened and why it matters.
