Innovative Tableau

Chapter 87. How to Make a Parameterized Scatter Plot

One of my favorite actionable or prescriptive tactics is to provide a scatter plot that the user can build themselves—even if they don’t know how to use Tableau! The parameterized scatter plot is very actionable, not only because the audience literally acts by creating its own analysis, but it often reveals new correlations and outliers that the business can act upon.

Scatter plots have several advantages: they’re able to show many data points at once, help illustrate correlations, and create a natural four-quadrant segmentation.

This chapter shows you how to make scatter plots even better by allowing your end user to choose the measure displayed on the y-axis, the measure displayed on the x-axis, and dimensional breakdown of the marks on the view.

How to Let Your Audience Build Their Own Scatter Plot in Tableau

For this tutorial, we rebuild the scatter plot from the Actionable dashboard in the Super Sample Superstore workbook. In the scatter plot on the right of the dashboard, the user can choose the measure on the y-axis, the measure on the x-axis, and the dimensional breakdown of the circle marks:

To allow an end user to build their own scatter plot, we use the tactic outlined in Chapter 64 in Practical Tableau (O’Reilly, 2018).

First, you must set up a parameter with a data type of String for each of the three choices ...