I think we’ve all been there… we are sharing a meaningful story we found in the data, only to have our end users get hung up on a previous peak or valley in our visualization. This often derails the conversation at hand and/or prevents our audience from hearing the rest of our message and/or reduces our chances of causing action.

One of the biggest challenges we face as data visualization practitioners is helping our end users avoid distraction. When our end users get distracted, it makes it more challenging to communicate the story in the data and our recommended actions. Ironically, one of the reasons users get distracted is because visualizing data makes it much easier to spot points of interest. Unfortunately, just because something may pique interest, it is not always relevant to the conversation.

This post shares an approach with the accompanying formula to do anomaly detection in Tableau. With anomaly detection, you’re able to focus in on the data points that matter and have a statistical explanation for your end users to help avoid distracting conversations.

Data normalization is the process of adjusting values from different scales to a common scale, providing a better “apples to apples” comparison of the values. For example, looking at varying metrics like ‘total cheeseburgers eaten per year by US state’, ‘total high fives given per year by US state’, and ‘total vacation days taken per year by US state’, will likely show that California leads the way in all three categories. This makes it sound like California is the best place to be in the US, and maybe it is, but these results are most likely due to California having the highest population (i.e. more people around to give each other high fives).

A better analysis would be to figure out how many high fives are given per person per year by US state. There are several ways to normalize data in Tableau including changing the aggregation of a measure, creating a calculated 100-point index, or setting a common baseline. There are also times when it’s valuable to see the raw, unnormalized numbers so our end users have the context of the original scales.

This post shares a method for (1) normalizing data and (2) providing you and your end users the ability to toggle the normalization on and off.

Timelines are not an out-of-the-box chart type in Tableau, but they can serve several practical purposes for your analyses and user experience. First, a timeline in Tableau can be used as a method for showing end users when notable events occurred in the business. For example, you can provide context by lining up a timeline of marketing promotions with a trend line to see when spikes align with your marketing efforts.

Second, a timeline can be used as a calendar showing upcoming dates of interest. In both use cases, you can use (1) a relative date filter in Tableau to dynamically display a subset of dates and (2) add dashboard actions to link to more information about notable events / dates on the timeline.

This tutorial shares how to make a timeline in Tableau and how to add an optional reference line to display the current day.

This chapter is excerpted from Practical Tableau: 100 Tips, Tutorials, and Strategies from a Tableau Zen Master published by O’Reilly Media Inc., 2016, ISBN: 978-1-4919-7724-8. Shop for Practical Tableau.

Seven years ago, I was fortunate to be introduced to Tableau, a tool for data visualization that has led to personal opportunities and contributions to my clients’ businesses that I didn’t imagine would be possible in such a short amount of time.

Even before that though, I was introduced to Google Analytics.

It’s no surprise then that some of my inspiration for the design, usability, and analytics of my data visualizations have been drawn from Google Analytics over the years. Sure, the Audience Overview report features an overly large pie chart for New vs. Returning visitors – a measure that is largely useless in an age of multiple devices, cleared cookies, and anonymous users – but Google tends to get a lot of things right.

What follows is a list of 10 Tableau data visualization best practices I use that I either consciously or subconsciously picked up from Google Analytics. In no particular order…

I’ve shared before that the reason I hitched my wagon to Tableau is the software’s flexibility. There simply is no other data visualization tool on the market today that allows me to create practically any visual or user experience I can imagine. Tableau is so flexible, in fact, that you can create web pages – and arguably web sites – all within the authoring and dashboarding interfaces.

This post shares a simple way to improve your end user’s experience by adding a button to a Tableau dashboard. These buttons can be used for several practical purposes including, but not limited to, (1) calls to action for a mainstream audience (2) a way to alert internal audiences of insights and (3) a method for linking related dashboards together.

Any image object on a Tableau dashboard can be made a “button” by setting its URL, but the approach shared here provides the added benefits of (1) showing a call to action on hover and (2) allowing you to format the button from right within Tableau.