9. Completing the Table Transformation


After making all the relevant decisions, this lesson will complete the third and final case study, replacing a large table of data with a more effective visualization.

To explore more Kubicle data literacy subjects, please refer to our full library.


  1. Lesson Goal (00:12)

    The goal of this lesson is to create the final version of our improved visualization.

  2. Adding Color and Shapes (00:20)

    Having decided what data should be encoded using color and shape, we can add the relevant data to our scatter plot visualization. In our case, using color for categories prevents us from using gradients for quantity, so we switch quantity to using size. It can happen that a change you make requires you to return to something you changed earlier and make further changes.

    In our case, the combination of nine colors and 4 shapes serves to make our chart far too complicated. Issues tend to arise when you use more than four traits from the visual hierarchy on a single chart, and in this case we have four traits on one chart.

  3. Considering Alternative Options (01:47)

    There are some options you can take that might help simplify your charts. We can swap the data used for color and shape for example. Generally, whichever data you want to emphasize most should be encoded using the simplest trait. In this case, color is simpler than shape.

    Another alternative is to replace generic shapes like circles and squares with icons relevant to the data. In our example, we could use icons representing the different products the company sells to represent data relating to those products. When you do this, it’s as if you’re not actually using the shape trait, which may make the chart easier to understand. In this case, it may be OK to have more than seven shapes, as the association between each shape and a product is intuitive and doesn’t require any thought.


In the previous lesson, we resolved the issue of how we want to complete the transformation of our table into an improved visualization.

Our goal in this lesson is to create the final version of our improved visualization.

Here, we can see the visualization we created earlier.

We decided that we would encode categories using color, and regions using shapes. Let's see what happens when we add categories using color. Using color for categories means we can't use gradients for quantity, so we've also switched back to size for quantity. Adding color makes it easy to see each category, where they're clustered, and where they're spread out. Even though there are nine colors, it's still easy to read. This is mainly because we don't have too many data points. Remember, try to shoot for the guideline of using at most seven colors. Regions is the only label left, and now we can finally assign that to shapes and be done.

Wow, the complexity really jumped up. This chart is now really bad. Even though region has only four shapes, the combination of size, color, and shape makes this so much more difficult to process. Just because you follow the hierarchy doesn't guarantee the outcome is going to be usable. Typically, using three of the traits is going to the limit. We made it to three, then adding shapes was just too much. Let's work through some options, and maybe we can find one that works.

Before, we decided to represent categories with color.

Here's what it looks like if we swap color and shape, it's not much better. Let's say that you absolutely had to use this chart or the last one. Think about what you're trying to help bring out and use the simplest trait for that. If regions is more important to the analysis, then assign that to color.

We have one more option to explore, so let's take a different approach to solve this. Instead of thinking of shapes in a simpler way like circles and triangles, use a shape that represents the data. Here, that's images of actual office supplies. Because you can make a natural association with the image, it's like you're not using the shape trait. There are plenty of resources on the internet where you can download these types of icons.

Also, something else to keep in mind is it's okay to have nine shapes, because the meaning is tied directly to the categories.

The rule of seven is more important when you have to keep the associations in your head.

This brings us to the end of our final case study. We've seen in this case how you step through the hierarchies, make trade-offs, and sometimes have to change previous decisions, depending on what you're trying to do.

In the next and final lesson of this course, we'll apply what we've learned from these case studies to a practical problem.