Did Netflix just kill its long-tail model?
Back in 2009, Chris Anderson posted this chart, showing how Netflix’s consumers were embracing the long tail of its offerings. As the number of movies in Netflix’s library grew from 4,500 to 18,000, the top 500 movies in the library went from constituting more than 70% of demand to less than 50% of demand.
This is the wonderful thing about online retailers: they can offer vastly more inventory than their local-store counterparts. And this is the wonderful thing, too, about Netflix streaming, at least as it existed until today. The big problem with Netflix, for many of us, before streaming came along, was always that we would stock up on highbrow ambitious fare which we knew we really ought to see, and then not be in the mood for something highbrow and ambitious when finally we had some spare time to watch a movie. With streaming, you could get the best of both worlds: a selection of fabulous rare DVDs, and instant access to something popular and brainless should you be so inclined.
Those days, however, are now over. You can buy the DVDs-by-mail service, or you can buy the online-streaming service, but to get them both you have to buy them separately: Netflix won’t give you even a single penny discount for getting both together rather than one or the other. In unscientific polls, more than a third of subscribers say they’re going to cancel; I doubt that anywhere near that many will follow through on that threat, but the fact is that Netflix offers much worse value going forward than it has done up until now. Streaming is great, but the selection is still extremely limited, and sometimes it doesn’t work at all for people with spotty internet connections. Having a few DVDs in familiar red envelopes was quite lovely, just in case there were problems with the internet, or you were about to go on a long journey with a laptop, or you needed to watch something only available at the end of Netflix’s long tail.
In shoving people out of their DVDs-by-mail schemes and into the streaming-only option, Netflix is reversing the trend seen in Anderson’s chart: the proportion of demand accounted for by its top 500 titles is almost certainly going to reach new all-time highs. At the margin, this move might well help encourage movie studios to allow their long-tail films to be available for streaming, but that process is going to take a long time and we might never have as much inventory available for streaming as we currently do in the DVD store.
Theoretically, of course, the long-tail model works even better with streaming than it does with physical DVDs, since all the physical problems associated with finding and managing the inventory of pieces of plastic go away, and filmmakers can upload their films for streaming even if they don’t have a DVD distribution model. But it’s going to take many years to get there, and in the mean time I think it would make sense for Netflix to allow its streaming customers to take out on physical DVDs any titles which aren’t available for streaming. That might give Netflix a bit less of a bully pulpit when it comes to negotiations with studios. But it would be much more consumer-friendly.
(Incidentally, is there a reliable number out there for the number of movies available on Netflix streaming? I’ve looked but can’t find one.)