Creating compelling and strategic dashboards in Datadog

Over the past five years, the amount of data we produce daily has increased dramatically. We create about 2.5 million bytes of data worldwide per day and the volume and complexity of the data grows exponentially every year. The rapidly evolving production of information makes it challenging to manage infrastructures struggling to keep up with the size.

Data is how we make sense of the world around us. We draw meaning, find patterns, and make associations based on our previous experiences. However, it takes an immense amount of human effort to sort through large volumes of complex data. We need to be able to visualize data in a digestible and compelling way without making it too simplistic. 

When working with millions of data points, we need to have an effective way to visualize what our business data is trying to communicate. Research shows the human brain processes visual inputs 60k times faster than text. The ability to process and exhibit even the most complex data in an efficient way is invaluable insight to a business’ ongoings.

At RapDev, we use Datadog’s Dashboards to make sense of large fleets, tracking various cloud service usage and ensuring the uptime of your apps across all regions. Dashboards help you analyze, compile and display data in a quick, easy-to-scan format with the most relevant information that is understandable at a glance on one single screen. 

Creating compelling and advantageous dashboards is both an art and a technical science. If you have an excellent way of collecting data but with no impressionable way to present it to stakeholders, then it's likely to not get the expected business outcome. We should strive for valuable data visualizations by making them strategically to display insights quickly and efficiently for end-users.This is the primary indication of orphaned records. An orphaned record will be visible in the list view but will be unable to be opened or selected in a reference field.

The lack of clear actionable data in a dashboard can be misleading. Even though we see some critical information about the above integration health and status, it might be better to add context to it with an overview and about section, notes or any relevant information that can help end-users make sense of the data they are presented with.

A good visualization should display data in an easy-to-scan format: This integration dashboard offers symmetry, concise descriptions and a well-formatted walk-through of its key performance metrics.
‍

Here, I’ve categorized several tricks I’ve used when creating dashboards:

• Use Screenboards if you wish to include images, graphs, and logs into your dashboards. They are great for storytelling views that update in real-time or represent fixed points in the past.
• Use Timeboards when representing a single point in time and you want to further troubleshoot, correlate and perform data exploration
• Use the Notes and Links widget to create headers and make it easier to parse rows
• Timeseries are a nice way to display the evolution of one or more metrics, log events or indexed spans over time.
• Use Line Graphs when reporting a single metric from a different scope.
• Stacked Graphs are great when you need to see the sum as well as the contribution of metrics in a single graph.
• Use Bar Graphs if you want to represent counts and show sparse metrics.
• Heat maps help you when reporting a single metric by a large number of groups. They show the distribution of values for a metric evolving over time. 
• Add markers to see ideal ranges for your metrics.  
• Distributions are a great idea when you want to show change in a specific time window or if you want to convey general health or status at a glance.
• Use Top Lists for ranking of hosts, clusters or any segment of your infrastructure by their metric values and if you want to easily spot outliers.
• You can create graphs using the JSON editor and have access to features not available in the GUI. 
• Change graphs are useful when you need to take two different timeframes as parameters and compare both to each other.
• Use Host Maps to get a higher level view of your infrastructure
• Query values can be great for extremely important metrics to see things at a glance without noise. You can have fun making use of conditional formatting/colors assignment to convey whether or not the value is in the expected range.
• Use Event Stream for monitoring footprints generated by your infrastructure and monitors
• Alert Graphs help you show the current information from any configured monitor or alert
• Check Status are a great way to illustrate the status for any configured check
• Apply built-in formulas and functions for more accurate results.
• iFrame widgets are especially useful when creating compelling dashboards because it gives you an instant look at your current home page. It lets you embed any iframe from another web page into a dashboard

At Rapdev, we use dashboards every day to analyze, monitor and troubleshoot business infrastructure. They provide a great way to stay on the same page with our teams and make data-driven decisions! Stay tuned for a follow up for this blog post, where we do a technical deep dive into how to build compelling dashboards.

If you are interested in implementing Datadog, reach out to us at chat@rapedv.io