Tableau Basics: Use Jitter to Spread Overlapping Points

Contributed by: Christopher Gardner

Difficulty level: Intermediate

When working in Tableau, you often want to see how items distribute over a range. For example, perhaps you want to see the range of profits for all your orders over the past few years. In Tableau, you would build a scatter plot by year to show the most common profit values for all your orders. The year of the Order Date would be on the Columns pane, the Profit measure would be on the Rows pane, and the Order ID would be added to Details on the Marks pane to split all the orders into unique points. The chart may look similar to the one shown.

Figure 0.

Looking at this scatter plot is a bit misleading. There are many orders each year that bring in matching or very similar profit values. As a result, the dots begin to overlap and cover each other up, hiding the true volume of orders. How would you go about uncovering these hidden points? How would you more accurately illustrate the volume of orders each year at each profit value? There are two ways to separate the points, also known as adding jitter, to make them more readable. Both involve simple calculations.

Random()

The first method to spread apart the values is to let Tableau assign a random value to each order and then plot that value within the year. This sounds complicated, but is really quite simple ...