I’m a full-time writer, but not, alas, a fabulously wealthy and/or successful one. James Cameron isn’t bugging me about film rights; Oprah isn’t plugging me on TV; fans aren’t lugging great stacks of my books around, chasing me for autographs.
It’s easy, when you’re one of the little guys in any creative field, be it fashion, books, movies or music, to envy the runaway successes and wonder what, for example, Stephenie Meyer’s got that you ain’t got. Are her books, objectively, truly so much better than everyone else’s? Or, more to the point, than mine?
Probably not, suggests recent research: in fact, runaway successes are runaway successes in part because they’re runaway successes…and efforts to figure out what “the next big thing” will be are largely wasted, because there’s no way to know.
That’s because people simply don’t make decisions as independently as we like to think.
A recent research project at Columbia University, led by Duncan Watts and Matthew Salganik, showed just how big an impact social influence can have on the popularity of something.
Through a website called Music Lab, the two registered more than 14,000 participants for their study. These participants were asked to listen to, rate and, if they chose, download songs by bands they had never heard of.
Some participants were only shown the names of the songs and bands. Others also saw how many times other participants had already downloaded the songs. Those who could see how often songs were downloaded were further split into eight separate “social-influence worlds”: they could only see the number of downloads a song received from other members of their “world.” This allowed the popularity of songs to evolve independently, eight times over.
If people made their choices completely independently, the scientists predicted, the most successful songs would draw about the same market share among both the participants who saw only band and song names and those who also saw how often the songs had been downloaded. As well, they predicted, the same songs, the “best songs,” would become hits in all eight social-influence worlds.
Instead, the most popular songs were much more popular, and the least popular songs less popular, in the social-influence worlds than in the independent group. Not only that, different songs became hits in each of the separate worlds.
This is where the idea of “cumulative advantage” comes in. Initially, all the songs were equal. But random choice by the participants soon meant that some songs were downloaded more than others. And once that happened, more participants started downloading them than the other songs, because they thought there must be a reason for their popularity–even though that popularity arose mostly by chance.
It may offer some slight solace to those who cling to their belief that they can’t be swayed by mass opinion that perceived quality did play some role in popularity. When downloads across all eight social-influence worlds were added together, songs the participants rated as higher in quality–“good” songs–had higher market share on average than “bad” ones. But the effect was miniscule. One song squarely in the middle of the quality rankings was number one in one social-influence world and number 40 in another one.
Or, as Watts put it in his New York Times article about his research, “A song in the Top 5 in terms of quality had only a 50-percent chance of finishing in the Top 5 of success.”
All of this indicates that things don’t become popular solely because they meet some previously unsuspected public desire or somehow match up with the public’s changing tastes. Instead, things become popular almost by chance, and then their very popularity changes the public’s taste. The market, in other words, influences itself.
Or, as the publisher of Lynne Truss’s bestselling book Eats, Shoots & Leaves put it when asked to explain its success, “it sold well because lots of people bought it.”
I’m not entirely convinced, so I’d like all my readers to help me conduct an experiment. I’d like each of you to go out and buy a dozen–better yet, two dozen–better yet, 100!–copies of my science fiction books Marseguro and Terra Insegura, just to see if we can artificially drive them to the top of the bestseller charts.
I’ll compile the royalties…um, I mean, the results…and report back just as soon as I can.
Well, if Oprah and Cameron will quit pestering me long enough, that is.