As marketers use data visualization to see and interact with their data more effectively, it’s most important that the information is simple enough to be understood. Complicated charts with multiple KPIs can be simplified for easier analysis and still adhere to best practice using Tableau’s parameters.
We’re often asked if we can look at multiple metrics in one visual. Suppose there are five KPIs shown on a chart for year-over-year comparison, all in one chart. It's not easy to read because we've tried to fit a lot in a small space. Instead of trying to fit all metrics into one visual or taking up a whole dashboard, analysts can leverage the Tableau parameter function to create a metric list and filter a visual to a specific metric.
Simplicity is Key
To avoid this, we've created a filter using parameters to select conversion rate, making it much easier to see what's been going on the last month, shown in Figure 1, below. Instead of showing all five metrics, we save real estate by using a parameter to change the visual’s level of detail. Another sophisticated function for using parameters in filtering a visual is to set up a list of dimensions to toggle between.
Figure 1:
Now, let’s look at Figure 1 a little closer. Opening a dashboard up to this view might be misleading because the y-axis intercept is not set to zero, it makes the year-over-year conversion rate look a lot more drastic than it really is. As a best practice, the intercept is set to zero, so new users aren’t alarmed by a seemingly drastic year-over-year change when they open a dashboard up.
One way we’ve solved for this is to create a filter to swap between visuals on the dashboard. In Figure 2, below, we have two separate graphs, one excluding the zero intercept and the other including the zero intercept. We can use a parameter to alternate between showing or hiding those sheets on a dashboard.
Figure 2:
We’ve now included a parameter that allows us to toggle between showing zero on the axis or hiding zero to drill into detail. But are we misrepresenting the data? Maybe. Check out the conversion rate (CVR) when we select “No” on the “Include Zero on Y Axis” toggle above.
Is CVR up or down overall? It’s hard to see the general trend with this view. While those that work closely with paid search know that day-to-day fluctuations are very normal, some may be alarmed at this volatile view and it could raise some unnecessary, panicked questions. You could look at a moving average, but the paid search team still wants to see the day-to-day data. How can you highlight the moving average, but still show the day-to-day?
This is where a dual axis trend chart is super handy. In Figure 3 below, we add a secondary axis to represent the five--day moving average of the CVR. The moving average helps smooth out the day-to-day conversion rate line, while the daily conversion rate is still represented by the dots in the chart. To make the five-day moving average line more pronounced, we can decrease the opacity of the dots.
And to make the dashboard more dynamic as you use it, you can replace the number in the moving average calculated field with an integer parameter. This allows for users to show a moving average across a longer timeframe for a longer-term analysis or adjust the parameter to zero if they want to see the KPI day by day.
Figure 3:
Bringing it all together
If you’re looking for a sophisticated way to fit a lot of analytics into a small amount of space, filters and parameters let you answer more questions without having to sort through more dashboards than necessary. We’ve shown you how to effectively use filters and parameters to do more with your limited dashboard real estate by using parameters to switch metrics, swap out visualizations, and dynamically change calculated fields. To learn more about how our Tableau Training program can help improve the skills of your analysts, click here.
Check out the other blogs in this series here.
