Stay Curious. Stay Connected.
MailNewsletters♡Donate Now
Sign up
*Password must be at least 8 characters and contain one uppercase letter, one lowercase letter, one number and one symbol.
Already have an account?
By submitting, you consent that you are at least 18 years of age and to receive information about MPR's or APMG entities' programs and offerings. The personally identifying information you provide will not be sold, shared, or used for purposes other than to communicate with you about MPR, APMG entities, and its sponsors. You may opt-out at any time clicking the unsubscribe link at the bottom of any email communication. View our Privacy Policy.
Sign in to your profile
Forgot Password?
New to MPR News?
Welcome back
to news
for you
Reset Password

Enter the username on file and we'll send you a code to reset your password.

Still require assistance or not sure what your User Name is?
Welcome back
to news
for you
Email Verification

A verification code has been emailed to

Still awaiting the code?
Tracy Mumford

Can a computer identify a bestselling book?

A woman uses a laptop
A woman uses a laptop computer.
Peter Macdiarmid | Getty Images 2007

Go Deeper.

Create an account or log in to save stories.

Log inCreate a free Account

Like this?

Thanks for liking this story! We have added it to a list of your favorite stories.

Show me!

Every week, The Thread tackles your book questions, big and small. Ask a question now.

This week's question: Can computers predict a bestseller?

They're sure going to try.

A company Inkitt bills itself as a "data-driven publisher." How does that work?

At inkitt.com, writers can post their original works and receive feedback from readers. Readers have free access to thousands of those new pieces across myriad genres. But how they read those works is being watched: Inkitt's algorithms reportedly track readers' behavior patterns.

Using those patterns, Inkitt claims it can identify potential hits and bestsellers. ("Read and fall in love with novels before they go mainstream," the site promises.) Inkitt then partners with traditional publishing houses to release the books. The first book identified in this manner will be Erin Swan's "Bright Star," a young adult novel scheduled to be published by Tor Books next summer, according to Digital Publishing News.

This method of using a test group of readers, pulled from the public at large, to weed through manuscripts isn't new. Amazon's Kindle Scout program works similarly. Inkitt's emphasis on data analysis and algorithms, however, is what sets it apart — they're not just crowd-sourcing the sorting of the slush pile, they're analyzing how the crowd does it.

Snapping up what's popular on the internet isn't new either, of course: Self-published hits that find an audience are often later picked up by traditional publishers, after they meet a measure of success.

Two authors with books on the Nielsen BookScan bestseller list last year started by posting their work on the internet. E.L. James, of the "Fifty Shades" empire, originally shared her work on a fan fiction site; and Andy Weir self-published "The Martian" before it became an award-winning film.

Inkitt's model aims to entice such writers to post their work on its site, and then analyze readers' behavior to find the hits. Will it work?

"Bright Star" comes out next summer — we'll see if it's a hit.