The Intertubes are alight this week with old news — that Netflix is the largest user of U.S. Internet bandwidth. Most stories cite a Sandvine report I won’t link to because you’d have to subscribe and I like you too much for that. Better still, look at the very interesting graphic above, courtesy of Arbor Networks. This chart has been floating around the net for a couple of months and shows the result of an Arbor study of several U.S. ISPs illustrating how we Americans spend our Internet bandwidth. There are three lessons I think we can learn from this chart: 1) that BitTorrent is no longer (or perhaps never was) the threat were were told by ISPs; 2) that video is by far the Big Kahuna of bandwidth, and: 3) that Netflix may be approaching the point where it is too big to fail.
First a look at BitTorrent, which ISPs love to complain about. Torrents are down to only eight percent of Internet traffic, but much more important is the fact that torrents have always been more polite than video streams. Here are two more graphs courtesy of Arbor Networks. First take a look at how web traffic varies over a typical 24 hour period: Now look at p2p traffic over the same period:The two are reciprocals of each other. This is by design, not coincidence. The nature of BitTorrent is to grab bandwidth not utilized by other services. So when web surfing declines in the late night and early morning hours BitTorrent increases.
Using only eight percent of Internet bandwidth and substantially less than that during peak hours, I think BitTorrent’s day as the Internet bogeyman are past, though I doubt the MPAA will see it that way.
Even more interesting is the rise of Internet video. Back in 2005 when iTunes users were downloading seven million three-minute music videos, readers of this column were downloading 2.5 million hours of NerdTV. I remember those downloads cost me $0.25 per gigabyte — ouch! In 2010 Netflix spent about $0.015 per gigabyte with an average 1.8-gigabyte movie download costing 2.7 cents to stream. Compare this to the average $1.00 Netflix spends to ship and receive every DVD and you can see their current business transformation from DVDs to streaming will lead to dramatically lower costs, freeing-up capital to buy more content. It’s a virtuous cycle that Netflix (and all it’s competitors to be sure) will attempt to leverage into its own form of too big to fail.
None of this is big news, I suppose, but think for a moment about the implications it has for both future services and for the commercial value of the Internet. Streaming costs are going down, not up, so what’s cheap today will be cheaper still tomorrow. These lower costs will allow higher quality (1080p video, for example) and they’ll shortly reach the point where stream costs will be lower than over-the-air broadcast costs on a per-viewer basis, which in the longer run is an inevitable prescription for the death of broadcast TV. It’s not a matter of if but when this will happen.
Even Luddites will be sucked into the Internet age if they want to communicate.
Despite having spent billions to help along the recent digital TV conversion, I’m sure the Federal Communications Commission will be happy to see broadcast TV disappear since it will do so with a flurry of spectrum auctions bringing-in many more billions to the Treasury. And that freed-up spectrum will go into more data services as we move toward the all-IP all the time future for carriers I have long predicted.
As for Netflix, it is hard to bet against the company. Hollywood studios glower and hint that Netflix will be deprived of content as current content deals — specifically Starz — expire, but that won’t happen. Dropping DVDs completely would transfer $2 billion straight to Netflix’s content acquisition budget through a combination of an increased subscriber base at lower prices and no more postal fees.
That $2 billion will buy a heck of a lot of crow in Hollywood, where cash is king.