Harvard Study Questions "Long Tail" Theory

mjasay writes "Remember 'the long tail?' That was the idea that there was gobs of money to be made in the more obscure tastes of any given market, enabled by the web. In recent research highlighted in the Harvard Business Review, however, the long tail theory comes under withering criticism. Not only is a hits-based business more profitable for vendors according to the new research, but the research suggests that consumers also derive more enjoyment from the hits, rather than the tail. In short, the researchers find that 'the tail is long and flat, and therefore that content providers will find it hard to profit much from it.'" Long Tail advocate Chris Anderson defends his theory, and it seems that most of the debate centers around how you define "head" and "tail."

Comments

I don't know about this.

[Paranatural]

Correct me if I am wrong, but the basic gist I got was that the 'long tail' theory states there is a lot of money to be made in 'niche' or 'subculture' elements, and the critics say this is wrong. I can wholeheartedly disagree. It's not easy to extract the profit from niche markets, and the long tail probably doesn't add up to the 'head' so to speak, in most cases, but there is certainly a lot of money to be made.

Take TV for instance. Your 'Heads' there may be ABC, CBS, NBC, and you could call things like The Discovery Channel, The History Channel, Comedy Central, Sci-Fi Network, etc the 'tail', as I understand it. And there is quite the potential for the 'tails', in this instance, to make even more than the 'heads'.

This is just as far as I understand it, mind you.

Re:I don't know about this.

[kklein]

I haven't read the book, but your explanation is how I understand the theory.

In the case of Amazon, I think your second conclusion, that Amazon does well on the long tail because they're the only ones there, is what is actually happening.

I buy a lot of esoteric testing, stats, and linguistics texts. That, in fact, is all I buy, and I buy a lot of them. And Amazon is the only place that carries them.

Same thing goes for CDs. If I want, say, the new NIN (ignoring for the time being that the last couple NIN releases I've actually gotten from his website), I go down to Tower Records (still alive here in Japan). If I want Freezepop, I need to go to Amazon.

Actually, in both of these cases, as long as I'm going to Amazon for the relatively unpopular thing, I might as well pick up the popular thing there while I'm at it.

I guess what I'm saying is that I don't think the long tail is necessarily a model for everyone, because if everyone did it, they'd see no returns. But for any particular market, a single long-tail retailer can make a bundle. Everyone jumps in and everyone loses money. It's a function of how big you are, not what you carry. Amazon basically has no competition on the "long tail" products, so they get 100% of that business. They have the luxury of doing that because they are frickin' huge.

So, all told, it just doesn't seem to me that the long tail is something just anyone can use, which is pretty much exactly what the study found.

The long tail most definitely exists in one area.

[juuri]

Porn. People do pay much higher sums for rather obscure or taboo things.

The problem with the "long tail" is that companies assumed it scaled. By definition it will only apply to a fringe. There is only space in such fringe areas for one or two dominant players; these players may make gobs of cash, but only in relation to their market size.

Of course large vendors aren't going to find it profitable to appeal to multiple fringe markets. The level of effort involved to support each individual small market is high and then combined with a number of markets means you end up burning through more manpower per dollar than a smaller dedicated company. It's the same problem of having too many products/SKUs/whatever, see DEC/Apple pre Jobs for an example of failing this way.

Re:Bullshit

[billcopc]

Precisely: their study is flawed, because they're basically saying that by dropping the obscure content that only 1% of their customers want, they're losing only 1% of their customer base (or less).

If they were examining a service that caters _specifically_ to that 1% niche, things would be much different. If your business involves selling purple spikey beanie babies, and you stop carrying the purple spikies, you go out of business. If your business involves selling ALL beanie babies, then you only lose the few weirdos who wanted *ONLY* the spikey ones, if and only if there is a competitor to fill in the void.

Redbox explained

[dazedNconfuzed]

Redbox is a vending-machine movie rental system. I walk into the Wal-Mart a mile from my house, and there's this big, well, red box sitting there. It operates on the opposite of the "long tail": it only has a couple dozen of the very latest movies, has many (but not infinite) copies of each, costs $1 per night (just swipe your credit card), keep 'em as long as you want (after a month, just keep it - for $29 you've already paid for it), rental & return is rediculously simple with none of the "video store" hassle.

Instead of having everything anyone might be looking for (the "long tail" model), it has a few things that most people will probably want (say, the "dirt cheap blockbuster" model). Turnover of content is very high, so there is most likely something sufficiently interesting (for a buck a night, that's a lot) there at any time. Content range is very narrow, so customers can browse very quickly; covers of most movies available are shown on the front of the vending machine, so one can review what's available in just a few seconds (a thourough list is available by touchscreen) even while someone else is actually using the machine. And with rentals being just a buck a night, getting something or keeping something a few days is trivially cheap.

It complements Netflix/Amazon thus: instead of getting exactly what I want in a few days, I get something satisfactory right now. My "long tail" providers can find anything I specifically want within a few days, but if I simply want a couple hours' entertainment now I can get something suitable, dirt cheap, in a few minutes. And when I take a rental back, it's just too easy to pick up another. It also fills the gap between "long tail" services and TV's "you'll watch what we want when we want" model.

That the box is located at the entrance to a store which thousands of people frequent with great regularity, rather than being a special trip, completes the winning business model.

Re:I don't understand

[emilper]

The long copyright term is not for the one-hit wonder bands, or the b-movie produced during the war to sell bonds. It's for the cream. It's for the Beatles. Cindella. The one in a thousand quality movie or band that people still think have meaning to them even now, 40, 50, 60 years later.

... or maybe it's for accounting: you own the copyright for a 2000 titles nobody bought for 50 years, but you can write them as being worth 2,000,000$ each, and then claim, in front of your shareholders, that the company is worth at least 4,000,000,000$. In fact, the company might be worth nothing, but who is to contradict you, since we all know that artists become famous and bring in money only after they die.

Re:I don't understand

[Gewalt]

... or maybe it's for accounting: you own the copyright for a 2000 titles nobody bought for 50 years, but you can write them as being worth 2,000,000$ each, and then claim, in front of your shareholders, that the company is worth at least 4,000,000,000$.

Then why not make those people/companies pay imaginary property tax on those imaginary properties?