In this module we will build on the analysis and dashboard you created in Module 2 (Enhance your dashboard). Now our dashboard looks great, but we need to add more interactivity to allow users to answer follow-up questions related to initial insights surfaced in the dashboard. Equally important is enabling users to take action directly from the dashboard by launching other applications or communicating with team members. We will explore concepts around adding interactivity such as cascading filters, filter actions, navigation actions, URL actions and alerts.
Cascading filters means limiting options shown in one filter or parameter control based on selections made in other controls. For instance limiting the list of cities based on the user’s selection in the country filter. Small touches like this save users time and makes the dashboard feel more responsive.
Filter actions allow users to filter other visuals or the whole dashboard depending on a selection made in a given visual. This is one of the easiest ways to enable users to ask more follow-up questions by drilling down and slicing the data by various dimensions used in the other visuals. While we applied a simple filter action in Module 1, in this module we will explore some of the customization options available for filter actions.
URL actions are used to launch other web pages when clicking on data point in a visual. They are important for that ‘last mile’ of analytics where I’ve found something interesting, and now I need to take action and do something about it. For instance if you are looking at project management or ticket data, URL actions allow users to launch the ticket management system and load the detail page for the ticket they clicked on. For sales data, they could allow users to launch the particular account or opportunity page directly in your CRM. Besides launching URLs which can be parameterized with data from your visual, you can also trigger emails to be sent, which is another common action users would want to take after finding something interesting (we also cover triggering emails in exercise 5).
Navigation actions are primarily designed to bring users to other sheets based on a selection, often times applying filters and drilling down in the process. A really common scenario is to take a user from a sheet with summary metrics to a sheet with more details upon a selection. Beyond this, there are other interesting experiences you can design using navigation actions to set a parameter value while keeping them on the same sheet. One example would be displaying an external web page inside the dashboard (using the Custom Visual Content type) and reloading the page based on selections made in a visual - similar to a URL action, but the page would be displayed inside the dashboard itself.
Alerts are notifications (in the form of emails) when the data indicates something needs your attention. More and more, users expect insights to find them and alerts one of the best ways to achieve this. They can be threshold-based, like ‘notify me if this number goes over 100’, or ML-driven, where QuickSight’s built-in anomaly detection algorithm can notify you if there were values significantly higher or lower than the expected/predicted amount. That being said, while alerts are great for telling you ‘something has happened that needs your attention’, that is just the beginning of the analysis and we need to use the rest of the techniques covered in this module and others to empower users to figure out ‘why that thing happened’.