“Tell me and I forget, teach me and I remember, involve me and I learn.” – Benjamin Franklin
I’ve always loved the above quote from Benjamin Franklin, and especially in the context of data visualization, because I think it applies to the practice in a couple of ways. First, simply telling a stakeholder your results is the least effective way to get your data-driven story to stick. By its nature, visualizing data provides additional context beyond verbally sharing findings, so practitioners are already a step ahead in communicating actionable insights. Second, tools like Tableau allow data designers / artists / developers to build in interactivity that lets end users find stories in the data on their own. When end users are involved, whether it be in the iterative process of a dashboard design or in interacting with a dashboard, the shared sense of ownership goes a long way towards making your data visualization a success. When end users find an insight on their own, they are more likely to remember it, and what’s better, do something about it – a holy grail of data visualization.
Allowing discovery is a tip I take personally because, as a huge sports fan in a relatively small market (Kansas City), my teams do not get as much coverage as some of the teams in larger cities on the coasts. With tools like Tableau, there is no longer an excuse to not include relevant information for every fan. This is something I keep in mind every time I create a new data visualization. This same principle applies in a corporate setting, as you now have the capability to provide relevant information to a variety of stakeholders in the same amount of space. For example, you can provide filters that change the view based on product categories or regions. You may have data size considerations, but it is theoretically possible to allow end users to look at every order or every customer individually.
In Tableau, there are three easy-to-implement ways to allow discovery.
We’ll start with the most obvious tool for allowing discovery: Quick Filters. Any filter used in the making of a view can be added to a dashboard. Quick Filters can be added in two ways:
- From within a sheet view, right-click a filter on the filters shelf and select “Show Quick Filter”. When the sheet is added to a dashboard, the quick filter will appear with the sheet.
- From within a dashboard view, (a) click the down arrow that appears when you hover over a sheet (b) hover over “Quick Filters” and (c) make the appropriate selection.
What Quick Filters look like:
Dashboard actions are a more subtle way to add interactivity to a dashboard. They also have the added benefit of saving processing time if your Tableau workbooks are published in the cloud. Dashboard actions are easy to create, but there are many options on how they can be utilized. The easiest way to implement dashboard actions is to click the down arrow that appears when you hover over a sheet and select “Use as Filter”. This will create a simple dashboard action behind the scenes that will filter your entire dashboard based on the item clicked in the sheet that you opted to use as a filter. What Dashboard Actions look like: Click a state to filter the trend.
Many people do not realize that Tableau workbooks embedded or shared in the cloud are interactive. This will improve as more and more people are exposed to the functionality of the software and end users become comfortable exploring a dashboard on their own. You may want to consider using hover actions so end users that are new to the software may stumble into the interaction capabilities that you’ve built in. For example, they may accidentally hover over the map in your dashboard, causing a change which makes them realize how to use your data visualization. This option is set when you build a dashboard interaction; instead of having the action execute on select (or click), choose to have the action execute on hover.
Parameters act as a wildcard that can be used in calculations. By showing the inputs for the parameters on your dashboards, end users can experiment with different scenarios on their own.
What Parameters look like:
Parameters are a slightly more complex way of allowing discovery. To learn how to build this simple what-if analysis, see our Tableau 201 post, How to Make a What-If Analysis Using Parameters.