Filters overview

When you create a Look, you can add filters to your new creation. These filters help specify the initial dataset for the Look, and can also be employed by users to examine the dataset in different ways. When someone accesses the Look directly, those filters appear at the very top of the window:

However, when you add that Look to a Dashboard (or when you create a Query Tile), these filters aren't available to users. Instead, users see only the visualization; they don’t see the filters (or the underlying dataset:

So does that mean that users are out of luck when it comes to filtering Dashboard data? As it turns out, the answer is no, they aren’t out of luck. Instead, Dashboard developers can add filters to the Dashboard itself, and then specify which tiles are affected by those filters and which ones aren’t.

The best way to explain what that all means is to look at an example. Here we have a Dashboard with two Looks, both of which deal with date-related data. We also have a single filter, which enables you to specify the year you want to view data for:

And then, just for fun, we’ve configured the filter so that it only works on the first Look on the Dashboard; the second Look will not be affected by the filter. In other words, suppose we set the filter to 2018. Here’s what the Dashboard looks like after we click Run:

If you look closely, you’ll see that the first Look, the one affected by the filter, has a new dataset: it only shows data from the year 2018. By comparison, the second Look is unchanged: it still displays the exact same dataset. Why? You got it: because we didn’t apply the filter to that Look. With Dashboard filters you can determine which tiles are affected when a user changes the filter and which ones aren’t.