April 22, 2018 | Written by: David Provan
Categorized: Data Analytics
Share this post:
This article was written by Gray Cannon, Analytics Consultant with IBM iX Events.
Anyone who has walked alongside the pristine fairways and blooming azaleas at Augusta National in early April will tell you the Masters Tournament is unlike any other event in professional sports. The mission of the Masters Digital team is to share this unique experience with the millions of golf fans who are not able to attend. Enabled by IBM as the official IT partner, the team brings new features each year to its digital platforms in hopes of delighting golf fans around the world.
But how do we know if the features hit the mark? IBM builds anonymous data collection into each of the Masters platforms in order to assess the usefulness of new features. As in any data-driven organization, the project stakeholders from IBM and Augusta National are eager to learn from usage patterns, so they can continually improve user experience. The Masters Digital program spans many platforms including web, mobile apps, TV apps, and wearables. The team considers high-level metrics such as users and visits as well as more focused statistics around features such as the leader board and live video. In order to consume all this information, flexible and interactive data dashboards are needed.
Cognos Dashboarding in Watson Studio
To create the needed dashboards, the team chose Cognos Dashboarding from IBM Watson Studio. The tool was well-suited for the task because the digital analytics data sets have several different shapes and require a variety of chart visualizations. A key element of the Masters dashboards is the ability to filter the visualizations during analysis. For example, the user can filter the charts for a single platform, such as the iPhone app, or filter for any specific combination of platforms, for instance, iPhone App + Android App.
For a weeklong annual sporting event, daily year-to-year comparison is an important technique, so tables with this information are used throughout the dashboards. However, it’s not enough. Over the course of a tournament day, stakeholders want to know the impact of star player’s round beginning or of a live video channel coming on air. For this kind of analysis, there are charts trended over the current day that update every 15 minutes. In Watson Studio, the team set these visualizations to refresh at regular intervals, so the data is always timely.
Using an IBM Cloud service called Cognos Dashboards Embedded, the Masters dashboards are embedded into an existing web application that houses other analytics services. Embedding the dashboards into this web app secures the dashboards using IBM (w3) authentication for IBMers and IBMid for other teammates. The analytics team appreciated this approach, as there were no user licenses to manage when adding new users.
To update the data sets, an IBM Cloud Functions action fires a request to an IBM Cloud Foundry Python Flask web service. These requests are sent according to a schedule set using time-based triggers in IBM Cloud Functions. Upon receiving each request, the web service retrieves data from the reporting APIs of the various digital analytics services. The web service then stores the data in IBM DB2 Warehouse on Cloud, and the data is immediately reflected in the dashboards.
A digital program with the scope and reach of the Masters Tournament creates a tremendous amount of data. Thoughtfully designed dashboards are essential for digital teams who aim to learn from this data. Watson Studio simplifies the task of creating rich data dashboards, and the IBM Cloud provides the tools necessary to create the underlying data sets.
The solution was developed by Gray Cannon (@Gray_Cannon) and Ryan Whitman (@fathead555 )