A diverging bar chart is a bar chart that has the marks for some dimension members pointing up or right, and the marks for other dimension members pointing in the opposite direction (down or left, respectively). What’s unique about a diverging bar chart is the marks flowing down or left do not necessarily represent negative values. The divergent line can represent zero, but it can also be used to simply separate the marks for two dimension members, to represent a goal, or – as often seen with survey data – to show the break between desired and undesired responses. The drawback to using diverging bar charts is that it’s not as easy to compare the values across dimension members as it is with a grouped bar chart. If, on the other hand, your primary objective is to compare the trend of each individual dimension member, a divergent bar chart is a good option. I also feel that this chart type helps declutter a grouped bar chart, making the data more engaging and easier to understand. In this post, we will reverse engineer my viz, 50 Years of AFC vs. NFC Matchups, to show you two different approaches to creating diverging bar charts in Tableau.
I’ve always believed that it is not enough to master the tactics of Tableau and data visualization if you neglect to put some strategic thought into your process. There are several strategic frameworks I follow to get to an effective result as efficiently as possible, but they all share a common thread: a focus on the audience. That’s why I start every engagement with my first of two “vital questions”: Who is the audience? The answer to this question largely informs what my Tableau product will eventually look like and getting the answer right tremendously improves the chances of my visualization being adopted and causing action. This post discusses the different types of internal and external audiences, tactics for handling audiences at different points on the visualization maturity spectrum, and the four personality types that you may find yourself ‘selling’ your dashboards to.
As a consultant, I have come across several occasions when my partners needed the ability to display their Tableau dashboard in multiple languages. For example, I’ve worked with a non-profit organization serving children in South America. The company’s offices are in the US and most employees speak English, but their field workers are in South America where Spanish and Portuguese are the primary languages. I also recently gave a presentation in Ottawa, Ontario, Canada where they have a similar challenge of needing to display English to some users, but French to others. I was even asked if it’s possible to create a bilingual dashboard in Tableau. According to Tableau, their software has been localized to Chinese, English, French, German, Japanese, Korean, (Brazilian) Portuguese, and Spanish. This means that navigation options will be translated, dates will show up in the appropriate format, and currency will have local formatting. What this post will show you is how to design entire dashboards in two (or more) different languages and allow your end users to choose which language is displayed throughout a workbook.
To draw a highlight table in Tableau, you need one or more dimensions and exactly one measure. This is the same criteria to draw a raw text table in Tableau, except with highlight tables, you’re limited to one measure instead of one or more measures. This one measure is what encodes the cells in the table by the preattentive attribute of color. It’s essentially a spreadsheet with colored cells. As I shared in my post, Why do you visualize data?, the highlight table is my favorite chart type for introducing the value of data visualization. I think it works well because most companies are still using spreadsheets for most of their reporting, and by converting a text table to a highlight table, the audience is forced to take advantage of the preattentive attribute of color. This kind of becomes a gateway to more complex visualizations. Highlight tables are already more engaging and effective than a text table / crosstab view, but this post aims to provide three more ways to make your highlight tables even better in Tableau.
In my post, Why do you visualize data?, I share my personal exercise for illustrating the benefits of data visualization. I first show a raw crosstab of data – similar to what most corporate reports still look like today – and ask the audience to answer the basic business question of identifying the highest or lowest number in the table. I then convert the crosstab to a highlight table by introducing the preattentive attribute of color, which reduces the time to insight, increases the accuracy of insights, and improves engagement. In the exercise, I take the highlight table a step further by only coloring the highest and lowest number in the view, further reducing the time to insight and increasing the accuracy of insights. This post shows you how to highlight the highest data point and lowest data point on a view using table calculations. This is as much about sharing some technical know how as it is about introducing the important concept of using Tableau to answer business questions automatically for you.