2.12. Scatter & Bubble Charts

2.12. Scatter Plots & Bubble Charts

Scatter Charts

Scatter charts are a great way to show the relationship between two different metrics and identify outliers. For example, you might plot transactions and average order value, broken down by US state. This can help you quickly spot states with high transactions and average order values or those with few transactions and high order values.

Scatter charts can be particularly useful when you want to focus your analysis on specific data points or clusters of outliers.

Bubble Charts

Bubble charts are similar to scatter charts, but they incorporate a third metric. This additional metric is represented by the size of the bubbles on the chart.

For example, you might use a bubble chart to display transactions, average order value, and e-commerce conversion rate, broken down by device category.

Bubble charts can provide valuable insights by allowing you to view the relationships between three different metrics at once. This can be helpful for identifying patterns or trends in your data.

Log Scales

When there's a significant diversity in values across an axis, you might consider using a log scale. Log scales can help you better visualize the differences between data points, especially when one or more points are much larger than the others.

However, it's important to note that log scales can be more difficult to read for those not familiar with them. Be sure to consider your audience's familiarity with log scales when deciding whether to use this type of visualization.

Examples of Scatter and Bubble Charts

E-commerce conversion rate, transactions, and average order value broken down by device category: This chart can verify your expectations about which devices have higher conversion rates and transactions.
Product revenue and quantity sold across different product categories: This chart can help you see which products are selling well and generating the most revenue, allowing you to prioritize marketing efforts and inventory management.