The following is the second part of a two-part post. To read Part 1, click here.
With the advent of digital entertainment, there’s a whole new set of metadata that can be collected about a book, and there’s no doubt that publishers would like to get hold of this information. But it doesn’t, I would argue, mean that they can use it to publish books.
I can absolutely see the value in optimising search terms for your book in order to be discovered in large and overcrowded online bookstores. Metadata will be incredibly valuable when it comes to marketing and selling. But that is still largely going to be the job of the retailer. Publishers have always been about content, and while I think most publishers would still love to see this kind of in-depth reading information, I don’t believe it is of real world use in discovering, judging and commissioning actual content.
It would be great if more data made for more successful books. But in the history of publishing, the vast majority of major hits have been relative outliers. There are just too many variables for this kind of information to be of actual use in predicting and discovering new stories.
The only way I can really see it being used is for ass-covering. I once heard a story about a film industry mogul talking about test screenings of new films. He said that the only reason they existed was so movie executives, when they were justifying the reasons behind their investment in an unsuccessful film, could point at the test audience results and say, “But it tested well.”
Analysing data and making predictions based on past behaviour have their place and their uses, but unless we start pumping the information directly into authors’ heads (where the stories actually come from), this metadata isn’t going to be the be all and end all of publishing. Not every author is a self-publisher, and those who aren’t are still going to need someone to discover, nurture and develop their talent, and metadata isn’t going to be of any use in that.