Five Ways to Balance Data and Design in Tableau without Graphic Design

Ryan Sleeper from Evolytics
on November 26, 2014

Ryan Sleeper is the Director of Data Visualization & Analysis at Evolytics in Kansas City, Missouri. Ryan is the founder and leader of the Kansas City Tableau User Group, a Tableau Conference speaker, and 2013 Tableau Iron Viz Champion.

This is an excerpt from tip twelve in a session Ryan presented at TC14, Balance Data and Design. For more data visualization tips from the series, see Data-Driven Storytelling: Tips from an Iron Viz Champion.

Five Ways to Balance Data and Design in Tableau without Graphic Design

Accurate and honest data is the core of every great data visualization. If you are able to leverage data visualization best practices to make that data easy to understand and act on, you are a huge step ahead of the game. If you have ever heard Tableau’s VP of Visual Analysis, Jock Mackinlay, comment on the topic, you will quickly understand how vital these aspects are to Tableau’s product development and the practice of data visualization as a whole. Undoubtedly, data is the kingpin when it comes to data visualization and best practices are kingpin-B.

That being said, without balancing the quality of your data with a quality design, your data visualization will never reach its full potential. Before dismissing design as an unnecessary – or worse, undoable - component of your data visualization, ask yourself why you visualize data. My answer is to find and communicate actionable insights. To make insights actionable, or provoke change, your data visualization needs to be seen by the most relevant and/or largest audience possible. Design is the key to reaching that audience.

I often talk about the concept of being remarkable, my favorite Seth Godin principle. Being remarkable means that your work is so quality / distinctive / compelling in some way that it moves your audience to remark about it. In today’s world, that means sharing your work on social media (if you find this post worthy of a remark, please see the share buttons above!). What it may also mean to you in the corporate world is that your work gets passed up the ladder – creating change, recognition, promotions … fame … and fortune! The latter may be a longshot, but it’s worth a shot.

To help illustrate how big of a difference design can make to a data visualization, I have stripped out most of the design elements from my most socially-viral work to date, Your Salary vs. a MLB Player’s Salary. I returned most of the fonts, colors, and formats to their default settings in Tableau, but I want to point out that I could have minimized the design even further by returning the view to its default layout and standardizing the font sizes.

In the viz, the end user can type in their own salary and compare it to the salary of any Major League Baseball player across several different statistics. The concept and data are interesting enough – if not disheartening - to stand on their own, but I credit the ‘remarkability’ of the dashboard to its design. Here is the exact same view with some simple design upgrades.

Click the image or here to view the 'designed' viz.

In addition to the true data purists who do not see the value in complementing data with an aesthetically-pleasing design, one reason I hear people give for not trying to balance data and design is that they have no graphic design experience. If you look closely at the ‘after’ image above, you will find that the only elements in the entire view that required any Adobe Photoshop or Adobe Illustrator talent are the icons that form the center of the dashboard’s donut charts. On top of that, I will let you in on the secret that those icons are from a stock photography file that I purchased from iStock. I simply recolored the icons to my liking and pasted them in the center of a circle to create the donut effect.

My point is: You can do this.

To help you get started, here are five ways you can balance data and design in Tableau without any graphic design experience:


There are countless articles on color’s impact on data visualization, and for good reason. The full possibilities of color are beyond the scope of this post, but I will share my two biggest tips for harnessing the power of color:

(1) Use simple color palettes with five or fewer alternatives. Of course, sometimes the variety of your dimensions will dictate how many colors your viz includes, but remember that every addition puts more stress on the user to efficiently decipher the story in your data (i.e. think looking back and forth at a color legend).

(2) Mute your colors. The technical definition of muting colors means that you are reducing their hue saturation. The practical definition is that you are making the colors less intense, making them more pleasing to look at. An easy way to do this is to add transparency to the colors.

If you would like to learn more about the theory and psychology of color, see Leveraging Color to Improve Your Data Visualization.


Tableau comes out of the box with 130 different fonts. If you thought that Trebuchet MS (Tableau’s default title font) and Arial (Tableau’s default non-title font) were your only options, then this tip is for you! Fonts are a powerful way to differentiate your dashboard, and they serve the additional purpose of communicating priorities to your end user.

There is no one-size-fits-all approach to font design, but I recommend picking out one or two fonts that you think look nice, fit the message of your dashboard, and most importantly, are easy to read. Experiment with different point sizes (i.e. 8pt, 10pt, 24pt) to break up your data visualization and prioritize its elements in a subtle way.

For font best practices, see Tableau’s article, Fonts: Can You Read This?


Making thoughtful layout choices can help guide your end user through the story of your data visualization even when you aren’t there to explain it. Paying special attention to the spacing of each dashboard component also helps ensure that you end up with a clean design.

One best practice in regard to layout is to place the highest-priority content towards the top and left of the view, and relatively lower-priority content towards the bottom and right of the view. I have also always followed a general rule of thumb to keep dashboard components – including titles, filters, and charts – to twelve or fewer.

For more on layout, see the most-read guest post of 2013, Kelly Martin’s Dashboard Design and Layout Best Practices.


Similar to layout, implementing good usability is one way you can help end users get the most from your data visualization, even when you don’t have the chance to walk them through it.

Make no mistake, usability is an important aspect of design: UX design. Add filters and dashboard actions that allow end users to find their own stories in the data. Inevitably, when a user understands how to use your dashboard and finds insights on their own, they are more likely to stick and be shared.

Tableau offers so many capabilities, one of the biggest challenges is communicating the interactive features available. Andy Cotgreave offers some great insight and tips on the topic in his article, How do you communicate that people can interact with your designs?


In this case, I saved the best for last. You’ve likely heard the expression, ‘the devil is in the detail’. Well, I’m here to tell you - so is the remarkability of a data visualization. Whenever I share a data visualization that doesn’t have any graphic design (as in actual graphics), but the audience thinks that it looks great, I credit the meticulous attention to detail in the dashboard. Many times this type of audience can’t quite put their finger on why the dashboard looks good, but I know the secret to my success is that the dashboard is more polished than other work they’ve seen.

Examples include reducing lines, softening gridlines, adding borders to marks, adding transparency to reveal overlapping marks, formatting filters, and potentially dozens more.

This attention to detail, combined with the four other ways mentioned to balance data and design – even without graphic design expertise – will help your data visualization achieve its full potential by reaching and engaging the most relevant and largest audience possible.