How to Create a Trellis Bar Chart in Sigma Computing

In this tutorial we will discuss how to create an essential data visualization in any analyst’s tool kit: the trellis chart. 

Before we get started, this tutorial assumes you already have an account with Sigma so head to this website: Free Trial with Sigma Computing to request one if you don’t already have one! We will be using sample data provided by Snowflake and included in every Sigma instance. 

What is Sigma Computing? 

Sigma Computing is a cloud-based Business Intelligence (BI) tool primarily used for data exploration and visualization. Sigma was designed to increase the speed to insights by utilizing Snowflake’s lightning-fast computing power coupled with the familiarity of spreadsheets. Sigma offers code-free and code-friendly data manipulation and visualization tools, making it easy to slice and dice your data in every conceivable combination.

What is a Trellis Chart? 

A trellis chart is a type of visual structure that helps you organize data into multiple dimensions. It’s similar to a cross-tab, but instead of laying out the data in two dimensions, trellis charts use three dimensions, which can be used to visualize complex relationships in a way that’s easy to understand. Trellis charts are often referred to as “small multiples” because they are like mini bar charts repeated across the page to help the end-user see the same metric across multiple slices.

When to Use a Trellis Chart?

Trellis charts are great solutions for showing the same snapshot in slightly different scenarios. For example, showing sales totals across years compared against multiple regions. With each scenario lined up close to each other, it is easy for the end-user to 1) consume large amounts of data quickly and 2) identify trends and changes over a wide range of scenarios. 

an example trellis chart in Sigma Comptuing

How to Make a Trellis Bar Chart in Sigma Computing

Step 1: Connect to Data

For this example, I will connect to Sigma’s built-in Plugs Electronic Sample data stored on Snowflake. For instructions on connecting Sigma to your Snowflake instance, check out this blog.

Steps for Adding a Data Table:

  1. Add an Element
  2. Select Data Elements > Table
  3. New > Table and Datasets
  4. Connections > Sigma Sample Database
  5. Examples > PLUGS_ELECTRONICS > PLUGS_ELECTRONICS_HANDS_ON_LAB_DATA
  6. Click Select (in bottom right corner)
adding a table in Sigma Computing

Connect to Data

Step 2: Add a Bar Chart

Now that we have data, we need to build a bar chart. Click Create Child Element in the top right corner, then select Visualization. 

creating a visualization using a child element in Sigma Computing

Step 3: Build the Bar Chart

Next, we need to add the data to the bar chart element. For this example, we want to show the total price by state. (For this example, I have hidden the other columns we won’t use.)

  1. Drag State to X-Axis
  2. Drage Price to the Y-Axis
  3. Change the bar orientation to display horizontally
  4. Click Trellis on the Element Properties pane
  5. Drag Region to Trellis Column
    1. Sigma defaults to trying to display every possible combination of State and Region. So we are shown empty slots for State and Region combinations that don’t exist in our data.
  6. Click the Element Format menu (paintbrush icon)
  7. Click Trellis to edit the Trellis formatting options
  8. Uncheck Share Y-Axis
adding region to the trellis column

Steps 1-5

unselecting the Y-axis

Steps 6-8

Step 4: Finish Formatting

There are several formatting options I chose to apply to help make the viz more appealing. Feel free to apply different settings as you see fit. 

  1. Sort Store State
    • Click the downward-facing arrow 
    • Click Sort > Sum of Price
    • Repeat
      • Sigma defaults to Smallest to Largest, and I want to show Largest to Smallest
  2. Color by Region
    • Duplicate the Store Region column
    • Click Color under Marks Section (if still on Trellis panel)
    • Drag the duplicated Store Region to Color
  3. Label Price Values
    • Click Element Format (paintbrush icon)
    • Expand Data Labels
    • Click to Apply Data Labels
    • Click Element Properties (combo chart icon)
    • Click the downward-facing arrow on Sum of Price (on X-Axis)
    • Select Format 
      • I chose a Custom Format of “$.4~s“
an example of a completed trellis chart tin Sigma Computing

Example of a completed Trellis Chart

Conclusion

Because Sigma is so fast and powerful when crunching millions of rows of data, it can sometimes be tricky to figure out how to display all the information. Trellis charts offer meaningful insights by quickly and easily comparing the same snapshot across multiple groupings or multiple time periods. 

Want to learn more about how Sigma can help your organization translate and visualize your data? Reach out to our expert consultants! 

FAQs

What if I uncheck Share X-Axis? 

Unchecking Shae the X-Axis will remove the synchronization of the axis range. For example, in the gif below, the axis goes from $0-$55M but unchecking it allows for the East region to max out at just over $30M.

Unchecked X-Axis

Which visualization types have Trellis options in Sigma?

You can choose a bar, line, area, scatter, donut, and pie chart. The visualizations that do not offer trellis are combo, box, a single value (KPI), funnel, gauge, and map options.

Accelerate and automate your data projects with the phData Toolkit

Data Coach is our premium analytics training program with one-on-one coaching from renowned experts.