How to Make Small Multiples is the second of a three-part post from “Use Corporate Chart Types”, tip nine in the Data-Driven Storytelling: Tips from an Iron Viz Champion series.

Small multiples are a group of charts or graphs that share the same axes and scales, which allows the user to compare trends across dimensions in a single view. They have been praised for their ability to provide a great deal of context, reducing the need for end users to ask the dreaded, “….So what?”.

The term small multiples was popularized by Edward Tufte, who puts it best in his book, Envisioning Information:

At the heart of quantitative reasoning is a single question: Compared to what?

While smaller series of small multiples can work well on an executive summary dashboard, we typically like to use them as a second layer in an analysis.

Our executive level view, or ‘first layer’, may provide higher level information about a particular measure, such as the overall sales trend and progress toward goal.

This is a ‘descriptive’ view of the data answering the question, “What is happening with sales?”

From here, we may provide an option to view sales across different dimensions and sub-categories as a series of small multiples in a second layer of the dashboard (often located away from the first view).

While small multiples is still a ‘descriptive’ view, it helps answer the question, “Compared to what?”

How to Make Small Multiples in Tableau

Small multiples can come in several different forms, but for the purposes of this tutorial, we are going to show you how to compare a single measure across two different dimensions.

We are using the Superstore sample data that comes packaged with every version of Tableau Desktop.

To get started, select two dimensions and a measure.

We have chosen the dimensions of ‘Customer Segment’ and ‘Continent’ and the measure of ‘Sales’ for the view to follow.

Step 1 – Place one dimension on the ‘Columns’ shelf and the other dimension and your measure on the ‘Rows’ shelf.

By default, Tableau has created a bar chart for you. By placing ‘Customer Segment’ on the ‘Rows’ shelf, performance for each customer segment is read left to right, in rows.

By placing ‘Continent’ on the ‘Columns’ shelf, each continent is represented vertically, in columns.

Small multiples may be a series of bar charts, but we still need to add an element of time to trend the ‘Sales’ measure.

Step 2 – Place a date field on the ‘Columns’ shelf.

Let’s pretend that we would like to evaluate the seasonality of sales to answer a question such as, “Is there a certain month that I can expect a spike in sales?”

For this type of analysis, we will use a discrete date field with a monthly aggregation, which will always show us the performance per distinct calendar month. By using discrete months, we know that we will always have twelve data points per small multiple, one for each calendar month (January, February, March, and so on).

The Superstore sample dataset contains four years of data, so if we used a continuous date field with a monthly aggregation, we would have up to 48 data points per small multiple (four years x 12 months = 48 points).

By right-clicking and dragging my ‘Order Date’ field onto the ‘Columns’ shelf, you are given an extra option to select the date aggregation (i.e. Year, Month, Week).

To get the view to look as it does below, choose the ‘MONTH’ option that was colored blue. The blue indicates that the date will be discrete.

Note that Australasia contained incomplete data, so excluded it from the view by right-clicking Australasia’s column header and selecting “Exclude”.

Step 3 – Format the small multiples to your preference.

From here, all that is left to do is format the view to your liking.

Formatting changes made to finalize this view include:

• Hid the field labels for columns by right-clicking on “Continent / Order Date”.
• Hid the field labels for rows by right-clicking on “Customer Segment”.
• Hid both the X and Y axes by right-clicking on them and deselecting “Show Header”.
  • You may choose to keep the axes in the view for additional context.
  • To keep the view as clean as possible, opt to provide this context in the form of a title, and provide additional information in the tooltips that appear when an end user hovers over different data points.
• Increased the font size in the headers from 8pt to 10pt.
• Softened the column and row dividers by using dotted lines instead of solid lines.

Final Considerations

In this tutorial, we have laid a strong foundation for using small multiples, but Tableau makes it easy to build on this view by customizing it in many different ways. For example, instead of using lines, you can change the mark type to bars, which are particularly useful for measures that can be positive or negative, or areas – which provide a beautiful design alternative.

W have also seen small multiple views that display year over year data nicely and Ben Jones of Tableau has shown how to create small multiple maps in Tableau. So get creative, and build small multiples that help provide the answer to “Compared to what?”.