A dashboard lets you organize multiple visualizations together, save them, and share them with other people. The ability to look at multiple visualizations has its own benefits. You can filter the data using some criteria and all visualizations will show the data filtered by the same criteria. This ability lets you uncover some powerful insights. It can also answer more complex questions.

Let us build a dashboard from the visualizations that we have created so far. Please click on the Dashboard tab from the left-hand-side navigation bar in Kibana. Click on the + Create new dashboard button to create a new dashboard.

Click on the Add link to add visualizations to your newly created dashboard. As you click, you will see all the visualizations we have built in a dropdown selection. You can add all the visualizations one by one and drag/resize to create a dashboard that suits your requirements.

Let us see what a dashboard may look like for the application that we are building:

With the dashboard, you can add filters by clicking on the Add filter link near the top-left corner of the dashboard. The selected filter will be applied to all the charts. 

The visualizations are interactive; for example, clicking on one of the pies of the donut charts will apply that filter globally. Let's see how this can be helpful.

When you click on the pie for 222 Broadway building in the donut chart at the bottom-right corner, you will see the filter for buildingName: "222 Broadway" added to the filters. This lets you see all of the data from the perspective of all the sensors in that building:

Let us delete that filter by hovering over the buildingName: "222 Broadway" filter by clicking on the trash icon. Next, we will try to interact with one of the line charts, that is, the Temperature at locations over time visualization.

As we observed earlier, there was a spike on December 1, 2017 at 15:00 IST. It is possible to zoom in to a particular time period by clicking, dragging, and drawing a rectangle around the time interval that we want to zoom in to within any line chart. In other words, just draw a rectangle around the spike, dragging your mouse while it is clicked. The result is that the time filter applied on the entire dashboard (which is displayed in the top-right corner) is changed.

Let's see whether we get any new insights from this simple operation to focus on that time period:

We uncover the following facts:

1. The temperature sensor at location O-201 (pink legend in fig-10.12) is steadily rising around this time. 
2. In the Coordinate Map visualization, you can see that the highlighted circle is red, compared to the other locations, which are yellow. This highlights that the location has an abnormally high temperature compared to the other locations.

Interacting with charts and applying different filters can provide powerful insights like the ones we just saw.

This concludes our application and demonstration of what we can do using the Elastic Stack components.