A Milestone of Innovation in Publishing
2019 is already a milestone year for Wattpad. Just a few weeks into the year, we announced Wattpad Books, our first-ever publishing division. That might surprise you, since Wattpad books have been on shelves all over the world for years.
But Wattpad Books is different. It’s taking everything we’ve built at Wattpad, and using it to innovate in important new ways.
Our elevator pitch is simple: Wattpad Books is a new, data-backed direct publishing division from Wattpad. Leveraging the billions of daily insights from our global community of 70 million story-lovers, Wattpad Books combines human editorial expertise and our Story DNA Machine Learning technology to identify the trends, voices, and stories that are the future of publishing. By elevating the stories of diverse communities around the world, Wattpad Books is creating a new space for writers and fans of every genre.
We bring something completely unique to publishing: an engaged global community, the most diverse set of writers on the planet, and the technology to find every type of hit imaginable.
Publishing books isn’t new to us. In fact, nearly 1,000 Wattpad stories have been turned into books, TV shows, films, or digital project already. But this is different. This is our own publishing division. We’ve seen Wattpad stories explode in popularity. And even though they haven’t hit bookshelves yet, Wattpad Books already have a massive readership, with passionate fans who have spent hundreds of millions of minutes with these stories. We’re giving people a new way to experience and discover the kinds of stories our 70 million users love.
In recent years we’ve made enormous advances in our Story DNA Machine Learning technology. We can find stories faster than ever before, and we’ve had great success in adapting Wattpad stories for TV and film.
Working with amazing partners at Macmillan in the US and Raincoast Books in Canada, Wattpad Books is a chance to do more of what we’ve already doing with partners around the world: using data to find great stories, helping diverse new writes find an audience, and turning.
Data, Not Popularity: How our technology works
Some people might assume that our data-based approach simply means we can find stories with the most reads. This is a popular misconception. Wattpad Books will publish not only popular stories on Wattpad, but also up-and-coming hits with exceptional potential.
Using machine learning, deep learning, and recurrent neural networks, our Story DNA Machine Learning technology allows us to better understand and discover stories among the more than 565 million uploads on Wattpad. Story DNA deconstructs stories into their elemental features, such as sentence structure, word use, and grammar employed to find the next best-seller.
We are then able to compare different types of stories at a massive scale. We can compare stories on Wattpad to other stories on Wattpad, and we also compare stories to public domain classics. This mix of genres and storytelling styles means we can find stories with huge commercial potential or grand literary ambitions. The result is a rich and complex system that can identify a wide range of stories, styles, and genres at a scale that has never been possible in human history.
Our technology allows to look beyond simple metrics like reads or reading time. These are important, but just some of the data points we can examine.
Consider this: there are more than half a billion stories on Wattpad, and people spend more than 22 billion minutes each month on the platform. We’re able to analyze data from every part of that user journey. We look at the different ways users interact with a story, including the number of reads attached to an individual story, reading time, commenting, shares, and more. Then there is growth and velocity for each of these things (how fast are these metrics growing), compared to everything else or compared to similar content.
Trends work in a similar way. We can see how people are tagging stories, which stories are trending, which tags are growing fastest, and find emerging tags to see trends organically take shape. Tags tend to generate associated tags so we see trends and genres splitting into new and diverse kinds of stories, which we can track and analyze. We can then look into the actual stories driving these trends for insights, tracking their popularity and comparing them against other content.
So we now see the types of stories emerging, how different elements within these genres are becoming more or less popular, then which stories are also becoming more or less popular.
Data tells a story. It tells us the story of how people are reading, talking about, and interacting with a work. It also tells us how a story relates to others within its own genre or other genres.
The principles behind this aren’t new. These types of analysis, understanding, and comparison are exactly what editors, university literature departments, and book clubs have been doing for ages. The difference is that we’re able to discover and understand stories that have blossomed on Wattpad and at a massive scale.
And editors are still critically important in this process. From half a billion story uploads, machines can help us find stories with exceptional potential. We can narrow the field to hundreds or thousands, but the judgement and expertise of editors remains a huge part of what we do. Once the machine identifies potential, an editor steps in to work with an author, honing a story and nurturing its potential into a finished product.
Disclaimer: This is to inform readers that the views, thoughts, and opinions expressed in the article belong solely to the author, and do not reflect the views of Amnet.
Copyright © 2020 Amnet. All rights reserved. No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the publisher, except in the case of brief quotations embodied in critical reviews and certain other non-commercial uses permitted by copyright law. For permission requests, write to John Purcell, Executive Editor- Amnet, addressed “Attention: Permissions” and email it to: [email protected]