Nathan Bransford, Author

Wednesday, April 6, 2016

Moneyball for book publishers

Way back in 2007, I wrote about what I called the "holy grail" of book publishing: the blockbuster detector. Inspired by both The Wisdom of Crowds and Moneyball, I wondered about the possibility of a tool that could identify undervalued markets and help predict future sales.

2007, mind you, was the era before the Kindle and iPad even existed. These, of course, opened up a world of possibility, where you could assess on a granular level where people stop reading, and distinguish between books that people buy and books that people actually read.

So naturally, now that it's 2016, there's a company dedicated to helping publishers detect which books might be hits or duds based on the reading habits of beta readers.

This is probably just the tip of the iceberg. Readers and writers, are you alarmed at the idea that your publisher could scale back your marketing budget if people stop reading past page 60 of your novel? Are you excited by the idea you could gain access to this type of data and help you revise?

What do you think?


Ian Hiatt said...

I'd always love more data on readers reactions, but using that data to make hard conclusions is dangerous. You can make statistics say just about anything you want, so saying that readers stopped reading because Reason X is foolish. Likewise, saying that many readers didn't finish book 2 doesn't mean book 3 doesn't have a market.

What's really scary about this prospect is that it may be treated as an experiment where two books score similarly and marketing/book deals are scaled down for one and not the other and conclusions drawn based on the results. There are far too many variables in book sales to consider any comparison to be a 'controlled experiment'.

The publishing industry can be mighty slow to change, so I hope this is one area they tread lightly.

Christopher Holliday said...

Nathan, thanks for making note of this article. It makes me wonder if Amazon and other companies with extensive ebook analytics will ever start applying this as a type of metric for determining perceived quality of a book, particularly indie published novels.

It seems like something that could outweigh even the review system, as not every reader writes a review, but every reader will have a distinct measured reading pattern with specific books. If 50% of readers drop out by page 50, does it go in the junk list?

I wonder if Kindle Unlimited titles give them that ability right now.

Rachael E Stout said...

I think it was only a matter of time. As the years progress, this will only get more advanced. Consider putting a book on you wish list? Wait! Our algorithm shows you are only 20% likley to read this book- try this instead! I think Amazon is getting tons of this data now to use later.

What really scares me as an unpublished author who is looking at going traditional: will this effect my ability to get published? Authors with proven track records already have the leg up, unless you have a house behind you it is hard to get that recognition- even if your writing is amazing.

Have fun reading, but remember big brother is watching!

Chris Bailey said...

Oh, Big Brother! Leave me out of this. My habits do not always reflect my enjoyment. If I've been asked to beta-read, I will ALWAYS read to the end. If I pick up a book because of the buzz, I may or may not get 50 pages deep before I put it down. If I'm reading an author I've enjoyed before, I nearly always read the whole book. Sometimes I start a book that's good but intense, and if I'm not in the mood I'll switch to a lighter book and come back to the heavy one later. If I knew I were influencing marketing dollars, I might get a little rebellious. Like the last time I got a Nielsen packet. I refused to turn the TV on for the entire week. Big Brother, back off.

Dick Grimm said...

Ultimately, book publishers are in the business to make a profit. They want to sell books. So of course they're going to use these kinds of analytics if they think it'll help them reach their goals and choose the books that are most likely to give readers a perceived satisfying reading experience.

However, that won't necessarily result in better books getting published. It'll only help more "readable" books get published. And it'll probably never predict which books will become cult hits like Interview with the Vampire which then spawned a string of immensely popular vampire novels which revived the genre.

This "moneyball" approach is not good for books in general, though I'm not sure it's bad for books in general either. It'll most likely result in more books modeled after James Patterson's formulaic approach. It'll cause more money to flow to breezy page turners that make a lot of readers feel like they were absorbed in a story without having ever confronted any weighty issues. But that's already happening anyway. Or maybe that's always been the case.

It's up to the authors to do their own analysis on what makes a book both great and commercially viable and then try to fit their voices into stories that meet those criteria.

I personally am not going to worry about hitting all the right analytics to make my novel a breakout bestseller. I'm just going to write stories I love which I think will appeal to enough people that I can keep on writing for those people without starving to death.

Jonathan Thomas Stratman said...

Some of the best selling, most-beloved books come at us out of nowhere. They "sparkle" in a way that is contrary to the critical mass of that moment's critical thinking. Analysis inevitably looks backward, is based on what has gone before. I think most of us hope that something we write will be uniquely forward-looking. We can all think of classics that were rejected repeatedly. We can all think of books we love that we would have said we weren't interested in at all. Still, people will always attempt to game a system, to find a shortcut, to maximize profit. Predict the midlist? Maybe. More than that? I have my doubts.

Charlotte Grubbs said...

I agree with Jonathan. People in publishing think that catering to readers' tastes is the best way to guarantee bestsellers, but the true breakout hits always seems to be a book readers never even knew they wanted until it was in their hands. They capture the zeitgeist, a shared moment in our culture that nobody knew existed until some lunatic put words to paper, and some other lunatic published it. Who thought teens were clamoring for sparkly vampires before Twilight? Or a novel narrated by Death before The Book Thief? And if we're so good at predicting what readers want, how come nobody cashed in on the "BDSM fanfic" or "astronaut trapped on mars" markets sooner?

I don't think data analytics is going to help find that breakout novel before it happens. I do think it'll end up reinforcing trends, quite possibly while shutting out more diverse stories and authors, and turning off readers who want the next hot thing, not reheated leftovers. Remember when every other YA was dystopian romance? Yeah. I don't particularly want to go back to that. Does anyone?

Chris said...

100% agree.
If I read a book all the way through or not is my business. I'll spread the word if I like or if I don't.
Since I rarely leave a book half read the assumption would be I love everything! That is not the case.
If e books feed information back to whoever, that's another reason not to touch them with a e-barge pole.

She Writes said...

I'm not excited about this. Readers are flighty. Time is elusive for some of us/them. But, what do I know? I have shelves of books. I love bookstores. I hold out for the printed pages of books. Most everything else, I read online.

Chris Phillips said...

I'm excited if the info is fairly public and accessible for cheap to writers. If it's a publisher's trade secret, then I wouldn't like it.

Related Posts with Thumbnails