Does Netflix Model have Legs?

One of the enduring lessons of new technology adoption is that the transition from older patterns to newer patterns can take quite a long time. One of the other lessons is that older ways of doing things sometimes do not fall before the new.

There was a time when TCP/IP was considered a transitional signaling method that would be replaced by the open systems interconnection model promulgated by the International Organization for Standardization. That never happened.

There was a time when Integrated Services Digital Network was seen as a replacement for time division multiplexing networks, on the way to a next generation network based on B-ISDN using asynchronous transfer mode. ATM proved a modest success, but has been eclipsed by IP.

Everybody "knows" that linear video will someday soon be eclipsed, possibly replaced by, Internet-delivered video, consumed on an on-demand basis. So far, that transition is proceeding slowly, which is typical for most technology substitutions.

What is more interesting: the extent to which the Netflix "DVD by mail" channel has more legs than people suspect. Though both the Blockbuster retail model and the Netflix models have been considered "toast" for some years, Netflix keeps defying expectations, even as it readies itself for the digital delivery shift.

Kaufman Bros. analyst Aaron Kessler has raised his rating on Netflix to "buy" from "hold," based at least in part on a survey of 700 Internet users, which found that 20 percent of all respondents that aren’t currently Netflix subscribers plan to sign up for DVD by mail service in the next five years. Repeat: "five years."

Kessler argues that, over the next five years, the country as a whole could reach near 20 percent Netflix subscription penetration.

The “DVD by mail life cycle may be longer than current thinking,” he says. “While we would agree that a large percentage of the DVD rental market will move to digital in the long term, our survey indicates that the current life cycle of physical rentals may be longer than people think."

Specifically, 68 percent of respondents indicated that “the ability to watch videos on the Internet versus renting from a physical store or by mail” is not important to them.

About 26 percent said online video was "somewhat" important and only six percent said it was " "important." Ultimately, much will hinge on how broadly content owners and ISPs support Internet delivery, as well as pricing models they choose to employ.

But it is worth noting how indifferent most users say they are to Internet-delivered on-demand programming compared to retail rental or DVD by mail alternatives.

In the communications and computing business, one always has a fundamental choice: substitute local processing for remote processing, and local storage for remote storage. In other words, one always can make an engineering trade off between communications and local processing and storage.

The "DVD by mail" alternative substitutes local processing for communications (physical media is an alternative to network delivery).

Online video substitututes communications for local resources, in the same way that "broadcast" video substitututes communications for local processing and storage.

Though the conventional wisdom is that users value and want on-demand video, Kessler's survey suggests otherwise. Users don't care as much as we sometimes think. Unlike news, sports and other "real time" content, where immediacy, in fact "shared experience" is important, movie content can be time shifted, place shifted or device shifted with little apparent downside.

That suggests demand for news, sports and other "real time" content will shift to online faster than "non-real time" pre-recorded video.

All of which is a reminder that we all have to be careful about assuming we know very much about what users actually want, how much they value various delivery and consumption channels and what they are willing to pay for experiences.

Netflix might prove to have a longer-lasting future than most of us now think. But that has been true for several years.

Might the same be true for other applications and services most of us assume are "toast?"