Netflix is not a part of the internet

The deal between Comcast and Netflix has been the subject of some thoughtful analysis (including some with a bit of snark) of its potential implications for the broader internet economy. But overall, the public discussion of the deal has generated more heat than light.
One source of confusion, which I touched on in my last post, is the assumption that Netflix must somehow be an injured party here — that it had “no recourse” but to “capitulate” to Comcast’s demands because the ISP had it “over a barrel.”
Netflix wanted this deal. It had been pushing for a direct interconnection with Comcast for more than a year. Comcast wanted a direct-connect deal with Netflix, too, just as it has commercial interconnection agreements with Amazon, Google, Facebook and other large content providers. The negotiation with Netflix was over the manner and terms of that direct connection. Like any consumer-facing online service, Netflix wanted as much control over its users’ experience, which means controlling as much of your internet routing as possible. Netflix may not have gotten everything it hoped for out if its negotiations with Comcast, but that’s business. If you don’t like the rules, don’t play.
Another big source of confusion about the deal is a misunderstanding of the business Netflix is actually in. Netflix is not an “internet” company. It’s business is not premised or contingent on the organization or underlying technology protocols of the internet the way Facebook’s business is, or Google’s. Netflix is in the business of producing, licensing and distributing movie and TV content, full stop.
It uses the internet as one of its platforms for delivering that content (the U.S. Postal Service is another), but moving bits around on the internet is merely a necessary step to engaging in its main business of getting content in front of consumers.
From a P&L point of view, the servers, data centers, exchange points and long-haul transit networks most of us think of as comprising “the internet” are simply cost centers to Netflix that add no value in themselves to the product.
Like any content distributors, Netflix is concerned with managing the necessary logistical pieces of its business as efficiently and cost-effectively as possible. As often as not, that means doing as much of its business as possible directly with as many consumer-facing end points as possible — as Netflix has been trying to do with ISPs generally for more than two years now.
As for whether the arrangement with Comcast means Netflix will now get preferential treatment of course it does, but so what? In the content business at least, big distributors always have preferential access to the market. That’s why indie record labels and movie studios sign with the majors for distribution. Access to the market is a function of scale and market share, and smaller distributors, by definition, don’t have those things, so they have a harder time getting their content in front of consumers.
That may seem wrong, or unfair on some level. But it’s not a new phenomenon in the content business and it certainly didn’t start with the Netflix-Comcast deal.
Sometimes, individual consumer-facing end points — retailers — gain unwonted leverage with distributors by virtue of their market share, which can harm distributors both large and small (although not necessarily consumers). How that market power was amassed, and whether it’s being used abusively are legitimate questions for regulators and policymakers. But they’re not new questions, or unique to the internet or to internet-based businesses. And they’re different questions from the ones too many people are asking about the Comcast-Netflix deal.
Netflix is in the content distribution business. Content distribution is one business (of many) that avail themselves of the internet for their purposes. Analyzing the strategic decisions those businesses make through the prism of the technology they use doesn’t always tell you much about the business.