March Madness 2015 was a success in terms of many metrics including ratings, ad revenue, audience size and live stream volume - all of which should come as little surprise as the event breaks records in these areas annually. But perhaps the most important stats are ‘missing’ in action.
The metric in question is perhaps the most important, given the rise of the event, to all stakeholders in live event streaming. The length of the average stream (duration) is what’s likely under wraps and could tell the industry the most about viewer engagement, especially when compared to numbers from past March Madness streaming.
March Madness streaming live has come a long way since the early days and has evolved over the years with technology and business models. On the infrastructure side alone – namely content distribution and quality – many far more recent streaming industry entrants could learn much from the annual multi-week college basketball tournament.
The Value of Visibility
Live streaming management, optimization and measurement packages from kingpin Conviva use these metrics to provide – in great detail – the numbers in question to content owners, broadcasters and service providers. The company’ customer list is a virtual ‘Who’s Who’ of the technology, media & entertainment (TME) industry.
And though Turner claims it doesn’t have numbers for the metric above, it’s likely the sports programming giant may just want to protect what most would consider proprietary information.
The NCAA last week publicized a series of stats (see below) that confirm what most already know: Streaming March Madness is very alive, doing very well and continues to show growth in virtually every category as the total amount of streamed video (hours watched and volume of live streams) continues to climb skyward for desktops and mobile devices.
During the tournament, fans could watch games via live streaming on TNT, TBS and truTV’s digital platforms, as well as participating TV provider websites, according to NCAA.com. The fast-broadening availability of the app, has helped the overall cause. NCAA March Madness Live launched from the following: ncaa.com/marchmadness, CBSSports.com, bleacherreport.com, CBS Sports App, Bleacher Report’s Team Stream, the App Store, the Amazon App Store, Windows Store and Google (News - Alert) Play.
Need to Know
Turner did provide cumulative numbers. It says it had a 58 percent increase in live streams and a 51 percent increase in time spent viewing versus the 2014 tournament. Based on this what was the average stream duration (time viewed)? And did it increase or decrease from last year? And greater detail by device type used would be priceless.
The average stream length tells stakeholders a great deal, even by itself. Generally speaking, live streaming sports 1.0 was the actual online streaming of events like March Madness, an option offered around the beginning of the 21st century. Version 2.0 was streaming with help from content delivery networks (CDNs) to deliver a higher quality viewing experience.
Version 3.0 was applying tools such as Conviva’s to optimize the viewer experience by controlling the streaming. Add in the ability to provide customers a boatload of important numbers for a growing list of metrics. Version 4.0 is applying authentication (TV subscribers entering a username and password for TV Everywhere, TVE) implementations, which includes access from additional device types.
The TVE Priority
Given this live event streaming evolution, the length of the average stream (average viewer duration) is more important than ever. Especially when you consider streaming problems for long-form content of late, borne out of larger than anticipated – and inadequately planned for – audiences.
The average length of stream per device type is also priceless data for pay-TV service providers who, despite defections from HBO, with HBO Now, have been looking to their TV Everywhere strategy for about 5 years to help keep subscribers from leaving for far cheaper and increasing more numerous TV alternatives such as OTT services.
This year, to view the streamed games for TBS, TNT, etc. you had to have a pay-TV service that included the channels carrying March Madness contests and then enter the related information (a process called authentication) to be able to watch the game online. The onetime $3.99 pay option was dropped a few years ago.
TVE is a customer retention strategy whereby pay TV subscribers can view the channels in their package on devices other than their TVs; namely online and on mobile devices. Live sports is - and has been - seen as the biggest, most powerful draw for TVE with the annual March Madness tourney the crown jewel event.
The average stream duration figures per device type are especially critical because of the DNA of March Madness, which unlike the NFL, for example, streams games played live during the work week, in addition to nights and weekends. A sizable contingent of fans has been known to watch the game online from work because of these fluctuating start times. And with the option to watch live game action on smartphones, fans don’t need to miss out on the tourney action.
Live streamers typically use innovative packages that enable them to optimize the delivery of video – often based on the amount of bandwidth available per viewer – that includes the ability to dynamically route around network congestion, blocked routes and much more. And these situations do exist, and not just with streaming March Madness.
In fact, about a year ago, Conviva released some stats on March Madness performance in 2013, claiming that it’s package made some 4 million corrective actions per day during the multi-week basketball tourney in an effort to optimize the fan’s viewing experience.
While it’s beyond safe to assume that March Madness streamers are hardcore fans, the missing stats would let all parties know what that translates into in terms of how long the average fan watched a stream and at what times of the day/night.
Building up those numbers should be the first job, given the tournament success in posting record viewership and stream growth numbers every year.
This information, augmented with additional data, should help answer questions as to why fans only stay engaged (tuned in) for certain amounts of time per device type. Better still would be taking action to extend this period and/or make the most of the viewing time as it stands.
The Bottom Line
While live streaming optimization packages can tell you about factors in the network that may have affected viewership – though they are designed to get around these one-time show-stopping issues – knowing the average stream duration per device type will tell you more about actual viewership and associated challenges.
There’s no question that the March Madness tourney has been a success in terms of the number of live streams and audience growth. But until the masses can see the numbers on average stream length cut as many ways as possible (and in historical context/comparison), how do we know what to build – and build on?
NOTE: For general stat lovers, here’s what the NCAA released last week.
From NCAA.com, as of April 7:
NCAA March Madness Live, managed by Turner Sports, set all-time records with 80.7 million live video streams and 17.8 million hours of live video consumption during this year’s NCAA Division I Men’s Basketball Championship. This year’s record-setting live video streams are up 17 percent over last year, with hours of live video consumption up 19 percent vs. 2014. Additionally, TV Everywhere usage soared during this year’s tournament, recording double-digit year-over-year increases across Turner’s websites and apps throughout the entire tournament.