Using Data to Reinvent How We Write

Robbie Allen
Oct 21, 2015 · 4 min read

With the rapid rate of innovation in virtually all areas of our lives, it’s surprising that the way we write remains largely unchanged. Sure, we use the keyboard instead of the quill. But our actual writing process–crafting a single narrative one word at time–would be familiar to Shakespeare, Shelley, and Salinger. Now there is something new.

We’ve just launched the public beta of our Wordsmith platform. Wordsmith is a new way to write and develop content using data. The process is part writing text and part writing logic, with data as the glue that ties everything together. Instead of writing a single story at a time, you create a story structure that can generate an unlimited number of articles. Wordsmith updates the writing process for the era of Big Data.

It’s an exciting time at Automated Insights because we are sharing parts of our platform that have automated earnings reports for The Associated Press, fantasy football recaps for Yahoo!, sales summaries for Allstate, and many others.

It’s Time for a New Way to Write

When it comes to writing, most innovations have centered on the distributionof content, not the creation of content. The printing press let publishers produce text in large volumes. The web and email enabled low cost digital content delivery. Finally, social platforms have created a “pull” environment where content finds interested users. Content distribution has come a long way.

In terms of the actual writing process, there has been far less innovation. Typewriters and word processors have been evolving slowly since the late 1800s, but even the most popular word processing software, Microsoft Word, hasn’t really changed the writing process in its 30-year history. Newer platforms like Medium have made it cool to get out of the way instead of giving writers new (or even traditional) capabilities.

On top of that, the availability of data should be a great asset in the writing process, but it largely goes unused. Big Data’s impact on writing has been hampered by the lack of integration with easy-to-use, widely accessible writing tools. As a result, many data-driven insights are hidden in complex charts and graphs that require additional — and manual — narrative explanations.

Why We Need to Create More Content

Automated Insights has been working on a platform to automate writing since 2007. In computer science terms, this is referred to as natural language generation (NLG). We call our platform Wordsmith. It can take a story structure and a dataset and generate numerous pieces of content that sound like a person wrote each one of them individually. For the last couple of years, Wordsmith has generated more content than any other company in the world. In fact, we create more content in a week than all the large media companies combined create in a whole year. Last year alone, Wordsmith generated over one billion pieces of content with a team of just 50 employees.

Some may argue that there is already too much content floating around. I agree. The problem is that it’s generic, untargeted content. At Ai, we focus on personalized content. Instead of writing one story and hoping a million people read it, Wordsmith can create a million stories targeted at each individual user and their preferences. It’s a story that is totally unique to each user because it is powered by their data. You can see a variety of examples on our website. We apply the same technique for large enterprises by creating individualized reports that go to everyone in a sales force, as we’ve done for Allstate, or monthly bill statements that go to every customer, as we’ve done for Comcast.

Early on it became apparent that instead of having only Ai employees configure Wordsmith, we needed to get it in the hands of more users. If a 50-person company can create more than a billion stories, what if anyone who wants to communicate insights around their data had access to the technology? I resisted opening up the platform earlier because it’s very difficult to create a new data-driven writing interface that is intuitive for business users. But if it was easy, someone else would already be doing it.

Thinking About Writing in a New Way

Writing fueled by data presents myriad opportunities for creative expression and driving business outcomes. There are a few prerequisites. You have to think about structuring your content to embrace variables. You have to understand your data at a basic level. Over time we will be releasing new tools and features that make data-driven writing easier and more automated. What you see today with Wordsmith is just the beginning.

When we first went to market with this technology in 2010, our goal was to get people to understand the possibilities. When I’m asked, “What are the limits of Wordsmith?” I often respond, “Your imagination.” With Wordsmith, the cost of content is an order of magnitude less than any other option. Meanwhile, the scale of creation is orders of magnitude greater. You have to use your imagination to explore what’s possible.

Be Part of the Beta

The Wordsmith Beta is now officially open. I invite you to request access today and start revolutionizing the way you write content. We’re excited to see how you put Wordsmith to use.

Originally posted on the Automated Insights Blog.

Robbie Allen

Written by

Co-Founder & CEO @InfiniaML, Founder @Ainsights

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade