Bandwidth for Over-the-Top Applications Comes at a Price – But, Who Pays?
July 29th, 2009
Alan Robin, CEO, Xangati
Over the past couple of months, I’ve been struck by the surge in use of over-the-top (OTT) applications, such as Internet video. By way of example, in March 2009, ComScore Video Metrix last estimated Netflix’s online viewership at 645,000 uniques per month and growing (in more recent interviews, Netflix now estimates online viewership to be in the millions, largely attributed to the growth in broadband). Additional ComScore data collected in March 2009 cited viewers watching nearly 7 million video streams in that month, with the average time spent watching per viewer being an amazing 2 or more hours – numbers that rival leader YouTube.
Netflix CEO, Reed Hastings, has been interviewed numerous times talking to the growing success of their business and specifically the economics of movie streaming (see http://tinyurl.com/lmxoe3 and http://tinyurl.com/kj4zys). However, in the context of those discussions, there is no talk of the network the online services actually run on. The fact is, Hastings relishes that it costs much less (a mere nickel at his estimation) to stream a movie over the Internet than it does to mail a DVD. The way he looks at it, Netflix is paying the studios for content rights instead of the Post Office for content distribution. But, the question remains: What are the economics for broadband providers?
Clearly, there are two sides of the story:
- The application guys follow the misperception that the network is free. Much the same as in the Netflix example, many content providers think the Internet is just awesome, high bandwidth is a given and you don’t have to pay for postage. But the reality is, it does cost more than the mere nickel Hastings claims for a 2 hour movie (e.g., connectivity to the house, capacity to the neighborhood, etc.) and the burden is simply just thrown elsewhere.
- Enter the broadband providers. Granted, they aren’t doing a good job of explaining their challenges and financial burdens in terms of network costs, unpredictable subscriber usage patterns, and rising support costs related to poor performance of OTT applications. Moreover, they are making things more difficult for themselves by also inadequately tying the above back to some of their new controversial business methods to stay in the black (e.g., usage-based billing). However, the fact remains that while companies like Netflix are saving money, broadband providers are seeing revenue streams flow off their network to someone else—in fact, many of them have spent many millions of dollars developing their own content under the umbrella of IPTV.
In other words, service providers care deeply about the business implications of OTT applications; however, they don’t have a mechanism to capture the data on the impact of these applications (which can also include VoIP services like Vonage) and explain the salient issues in a way that is easily digestible and acted upon. Thus, a great void is created in having a framework for even opening up discussions with the content providers.
All of this further underscores the importance of educating the market—first to the OTT application providers and simultaneously to their customers. And based upon what we have seen thus far, it points to service providers needing better ways to data mine their subscriber application activity to generate the necessary content for the education steps that lie ahead. Clearly, usage-based billing (a.k.a. usage caps) alone is not the solution as Matt Davis reported in an IDC survey conducted in August 2008. That survey showed that four-fifths of consumers don’t approve of “bandwidth capping” and “light users are less likely to know what a GB is and more inclined to churn than heavy users.”
In order to solve the problem in a real and lasting way, the content providers and telecom guys need to start speaking the same language and come to the table for a discussion of the realities of consumer behavior today, how to best deliver and figure out a way to share the costs for making it happen. Without bandwidth, there’s nothing; so, the missing link seems to be: How much should network providers be paid? We are interested in your feedback.