Scatter Plot and Bubble Chart - Definition, Examples, and Best Practices

What is a scatter plot?

A scatter plot places individual data points on an x-y axis to reveal relationships between two variables. Each point represents one observation - its position on the horizontal axis shows one variable, its position on the vertical axis shows another. Unlike a line graph, no lines connect the points. The pattern of how the dots cluster, spread, or align tells the story.

Scatter plots are the primary tool for exploring correlation - whether two variables tend to move together, move in opposite directions, or have no relationship at all.

What is a bubble chart?

A bubble chart extends the scatter plot by adding a third variable: the size of each dot. Larger bubbles represent higher values of the third variable. This turns a two-dimensional comparison into a three-dimensional one, allowing you to see relationships between three measures simultaneously.

When to use scatter plots and bubble charts

These charts work best when you have a set of individual observations and you want to explore how two or three variables relate to each other. Typical use cases:

• Customer analysis - plotting acquisition cost (x) against lifetime value (y) to identify your most efficient channels. Add bubble size for customer volume to see the full picture
• Competitive landscape - mapping competitors by growth rate (x) and margin (y), with bubble size representing revenue. A classic strategy consulting framework
• Portfolio evaluation - plotting investments or business units by risk (x) and return (y), with bubble size for capital deployed
• Operational analysis - showing the relationship between production volume and defect rate across factory locations
• Correlation exploration - any time you're asking "do these two things move together?" a scatter plot is the starting point

How to read a scatter plot

Look for the overall pattern first. Do the points trend upward (positive correlation), downward (negative correlation), or show no pattern (no correlation)? Then look for outliers - data points that sit far from the main cluster often contain the most interesting insights. In a bubble chart, look for large bubbles in unexpected positions.

Best practices

• Label your axes clearly. Without labels, a scatter plot is meaningless - the reader needs to know what each dimension represents
• Highlight outliers or key data points with labels or a different color. In a competitive landscape chart, you'll want to call out your own company and key competitors
• Use quadrant lines where appropriate. Dividing the chart into four quadrants (e.g., high growth / high margin, high growth / low margin, etc.) turns an exploration tool into a strategic framework
• For bubble charts, include a size legend so the reader can estimate values. Without it, bubble sizes are ambiguous
• Don't overload the chart with too many data points. If you have hundreds of observations, consider using color or opacity to show density rather than plotting every point with a label