Hits vs. Blockbusters

A colleague rightly chides me for muddying the waters over the whole future of the hit thing.

First, I complain that the number one misinterpretation of the Long Tail is that hits are dead. As I explained in this post ("Hits aren't dead. Hits aren't dead. Hits..."):

What is dead is the monopoly of the hit. For too long hits or products intended to be hits have had the stage to themselves, because only hit-centric companies had access to the retail channel and the retail channel only had room for best-sellers. But now blockbusters must share the stage with a million niche products, and this will lead to a very different marketplace.

Then last week, in my update on the latest music sales, I start the post by saying "...it's time to start updating the death-of-the-blockbuster stats." So which is it, Chris? Are hits dead or not?

Well, first, I'm going to claim a little literary licence with that whole "death" thing. I call that rhetorical device "endism" and I half-heartedly defend it here. It's exaggeration, of course, but with a point.

But there's actually a real distinction I'm drawing between hits and blockbusters.  The reason I get so annoyed when people misread The Long Tail to think that I'm arguing that hits are dead is that the canonical powerlaw that defines the Long Tail requires hits. That curve is the shape of radical inequality, where a few things sell a lot and a lot of things sell a little. The right side of the curve--the long tail--can't exist without the left side of the curve--the short head--which is made up of hits.

But as markets vastly increase their available variety and the tools for consumers to find it, demand shifts down the tail, from hits to niche products. You still have radical inequality between the two, but it's not as pronounced as it is in traditional limited-choice markets. So the hits get a little less "hit-ey"--their sales decline a bit, as the aggregate sales of millions of niche products increases. In short, blockbusters turn into regular old hits.

So that's the distinction I'm making. Blockbusters--megahits--are dying in industries such as music (where the line of the moment is that "gold is the new platinum"), but hits will be with us forever.

You can visualize it like this:

Comments

[Kevin Kelly]

Chris,

I'm glad you have begun to clarified this. You seem to be saying that the new regime CHANGES the shape of the distribution curve. According to the model graph you drew, the new tail is longer than the old tail, and thicker, too. That is, works of a particular sales rank are now elevated in their sales, higher on the graph. Not into hit territory, but higher. Whereas before maybe they sold one or two, now maybe they sell a hundred. All this would be wonderful news if you had charts like this generated by actual data. I mean ones that showed how the curve CHANGES shape as an industry moves into this new territory.

Your cover "long tail" curve was a cannonical powerlaw curve. It was never clear to me whether you were claiming this ideal curve as the Old curve that we are moving away from, or the New curve we are moving into. If I am reading you correctly, you are saying the cannonical Powerlaw curve is what we are moving away from, and we are moving towards a different shape curve. So what is the math of that curve?

But maybe you are saying that the past was not a powerlaw curve, and they the new media regime IS a powerlaw curve? What then was the old equation?

Either way, you seem to be saying that the Powerlaw curve is not universal -- which is sort of heretical. Or that there is more than one powerlaw curve (which is what I suspect). In any case further clarification would be helpful. Even better would be actual data showing that in the past -- say 20 years ago -- books/music/movies followed this X shape curve, whereas now they follow this Y one -- which is governed by a slightly different equation.

[YLlama]

To me, it looks like the above are *both* powerlaw curves, but with different exponents.

[Ian Betteridge]

I'm glad you've begun to clarify this, and I agree with Kevin when he asks for data. I suspect that, actually, when you start looking closely you'll find that media markets have followed a curve that's similar to your putative modern one, and not your proposed old one, for a long time.

Case in point: Music in the 1960's must have looked like your short-tail curve, right? Wrong - the 60's saw many, many tiny record companies that issues niche records. Take the flourishing of gay music that was recorded for a gay audience and distributed through very small-scale means (see the excellent "Queer Noise 61-74" CD for a great history).

Of course, as the cost of the means of production comes down, the length of the tail will get longer, but that's been economic knowledge since Marx's time.

[Rupert Watson]

Looks like the UK is going to get its first unsigned band in the Top 40 chart.
http://news.bbc.co.uk/1/hi/entertainment/6248535.stm

[Kevin Marks]

The cover has to be schematic - if you draw it accurately, you get a letter L (if you square up the axes) or a long horizontal line with a tiny bump on the left end (if you keep them proportionate). The challenge is always to explain this withought having to explain log-log plots first. My efforts a while back didn't manage it

The net extends the range of the power law distribution.
If you look at relative popularity on the web, using something like Technorati, you get a power law curve that goes all the way down smoothly, to the bottom where you see pages that got just a single link.
If you look at popularity in the publishing world - movies, chart music or books - the curve starts out with a power law, but soon drops like a stone.
That's because in order to get a movie made, a recording contract or a book published, you have to convince somebody that you're going to sell a million tickets, a hundred thousand CDs or tens of thousands of books.
You end up in a zero-sum game, where people pour enormous resources into being number one, because number two is only half as good. The promise of the net is that the power of all those little links can outweigh the power of the top ten.

So it is encouraging to see Chris keep trying new ways to do so.

[Pablo Volenski]

For ever ?
that's the only that bothers me from all the excelents affirmations you make on this post.
It it's true: hits've got a huge industrial background and supporting, enormous enterprises that will get till the end in the fight of saving their business.
Anyways, I'm still believing that as in other moments of the human history, social revolutions (taking participation and Web2.0 as a social revolution) always mold the market, and not the other way arround.
I'm still thinking that the "always" you put in this post, it's really a "for some more time".
Thanks you a lot !

[Pablo Volenski]

For ever ?
that's the only that bothers me from all the excelents affirmations you make on this post.
It it's true: hits've got a huge industrial background and supporting, enormous enterprises that will get till the end in the fight of saving their business.
Anyways, I'm still believing that as in other moments of the human history, social revolutions (taking participation and Web2.0 as a social revolution) always mold the market, and not the other way arround.
I'm still thinking that the "always" you put in this post, it's really a "for some more time".
Thanks you a lot !