The US Open Puts Data Center Court
Source: US Open
The hallways beneath the Billie Jean King National Tennis Center in Flushing, NY—host site of the annual US Open tournament—have all the harmony of rush hour in Manhattan. United States Tennis Association (USTA) staff bull their way through the passages and the ball boys in their Polo Ralph Lauren blues—I swear they've got that horse logo on some sort of growth hormone—march single file toward center court. Stop moving, get pancaked from behind.
But as much as the US Open runs on human sweat and activity, it also relies heavily on the movement of extensive and granular data sets—from athlete and match information pumped to media and broadcast partners to fan engagement stats that let US Open staff know how exactly to allocate their digital resources.
The room where much of this takes place is quieter by comparison to the rest of the US Open premises—a modest cinder block cell with monitor banks and stats from last year's matches painted on the wall: "Of the 127 matches played in the men's singles championship, nearly 20% went to a decisive 5th set." Or: "Serena Williams lost serves just 6 times in the 7 matches en route to the women's singles championship."
It's this obsessive granularity and attention to detail that by necessity characterizes the USTA's digital strategy. The organization, which extended by four years its 23-year partnership with IBM, operates like a small business throughout much of the year. But when the US Open hits at summer's end, the chatter, online-content searchers, ticket buyers, and schedule seekers spike—meaning, the entire USTA is suddenly buffeted by a lot of information in a short amount of time.
Its data variety and volume is fired with the blistering velocity of a tennis serve. In other words, it's the three V's that most analysts use to characterize Big Data. Add a populace that demands real-time mobile content—the flashpoint occurred, according to USTA Director of Digital Strategy and Partnerships Nicole Jeter West, just last year—and the US Open suddenly finds itself a digital event as much as it is a live one.
Building a data stadium
The infrastructure that the USTA and IBM use to support the US Open has been steadily evolving since 1990. From a website to a basic mobile app, the current technology now supports platforms on mobile handsets, tablets, and of course the branded website. From an architectural standpoint, the USTA's digital presence migrated last year to a private cloud, spread across three virtualized service delivery locations (other tenants on the cloud include the three other major tennis events that constitute the Grand Slam tournaments, the Tony Awards, and IBM itself).
The move to a private cloud was big for Jeter West because USTA needed scalability and flexibility—especially since the organization posts up-to-the-moment analysis and content using technology from IBM's Smarter Commerce stack. For instance, the SlamTracker provides real-time statistical analysis on player performance; it also uses predictive analytics to examine eight years of Grand Slam Tennis data to find patterns around players, and applies this information against each player's opponents to determine the keys to a victory.
The problem is that is a lot of content needs to be delivered with near immediacy. (Note: Fans actually at the Billie Jean King Center will find that Time Warner's “high-speed internet” transfers information with the speed of a dead carrier pigeon). Most institutions can get by with a better-late-than-never approach. But at the US Open, there are no do-overs for whiffing.
“For us, there is no recovery time,” Jeter West says. “We have to have it all now or else it's missed.”
The move to the private cloud, and tying it into social and mobile analytics, allows the USTA to better allocate resources to accomplish this. For instance, if an upcoming match features a much-hyped player, the USTA will note an increase in social chatter, determine the player's popularity, and predict the upcoming schedule. Through this, the USTA can begin planning exactly what it will need in order to serve the right content, information, and analytics to customers.
Eyes on the ball
Planning for each US Open begins at the end of the previous year's tournament, when Jeter West's team and IBM look at the data and analytics to determine growth points and areas that can be improved. Last year was, as everybody knows, the Year of Mobile.
And this changed significantly the way people consume US Open information. Traditionally, the high points had come midday on weekdays, likely during those workplace-related procrastination binges that include Facebook, Reddit, and gossip blogs. But mobile has stimulated significant growth in weekend viewership when consumers glance down from the church pews to check out the latest updates.
Tracking this shift led to monetization opportunities that wouldn't have otherwise existed. Last year Jeter West noted there had been significant lift in the US Open's branded tablet and mobile handset apps. Moreover, she was able to see exactly where on those apps audiences were spending time—typically around scores and schedules.
There were proof points that the USTA could take to sponsors: We know where the eyeballs are. And this presented monetization opportunities that hadn't previously existed. For instance, USTA developed TrendCast for American Express, a sponsored addition to the US Open app that incorporated trending topics and original radio programming.
Game. Set. Match.
The US Open will continue until September 9. And there will likely be upsets, like the stunner that Haitian-America Victoria Duval leveled against 2011 US Open champ Samantha Stosur. But if Jeter West has her way, those unexpected surprises will be relegated to the courts.
“The cloud, for me, is why I get to sleep at night,” she says. “To know the amount of redundancy that takes place in the cloud…allows us to focus on the details: deliver the content every day knowing that the infrastructure is supported. That is really what allows us to scale and reach the amount of fans that we have.”