Textio, A Startup That Analyzes Text Performance, Raises $8M

Textio CEO Kieran Snyder took a quantitative approach to how language worked in her linguistics studies. And when she and her co-founder Jensen Harris were leaving Microsoft to start a new company, it was only natural that it would be centered around language in some way.

That’s how Textio, a startup that analyzes text for how well words and phrases perform in certain scenarios, was born. The company today said it raised $8 million in a financing round led by Emergence Capital. Cowboy Ventures, Bloomberg Beta, and Upside Partnership also participated in the financing round.

“We had this premise that word processing in text hadn’t been disrupted in a while, from command line to GUI,” CEO Kieran Snyder said. “We had the internet come along, it was about social and sharing, and we think that AI and the set of related technologies is the next big disruptor of text. If you know the performance of a document before it’s ever published then you can fix it before it’s published.”

Textio’s first tool looks at talent acquisition documents — like job postings — to determine how well they will perform among candidates. Certain words and layouts attract more candidates than others, Snyder found, and those predictive analytics are baked into the service. For example, Textio shows that job postings with bullet points tend to perform better than job postings without them.

Right now it’s used for talent acquisition documents, but it’s pretty easy to see that the technology can be applied to documents that include common phrases — such as email, resumes, or other kinds of messages. If the technology works, it can theoretically begin building up scores for those kinds of documents, which is likely what attracted investors to the product and the team.

Another reason it might be so valuable to investors? Its customers. Already Textio is being used by companies like Twitter, Atlassian, Starbucks, Square and Microsoft. Natural Language Processing technology has very broad applications if done right, which makes it an attractive bet for many investors.

Textio recognizes more than 60,000 phrases with its predictive technology, Snyder said, and that data set is changing constantly as it continues to operate. It looks at how words are put together — such as how verb dense a phrase is — and at other syntax-related properties the document may have. All that put together results in a score for the document, based on how likely it is to succeed in whatever the writer set out to do.

Given who’s likely using Textio, it’s important that it feels easy to use — hence the highlighting and dropdown boxes rather than readouts. Snyder said, at its core, Textio can’t feel like a statistics tool, and that’s probably because the kinds of people using it might not always be NLP experts.

Of course, there are potential competitors in the space when it comes to natural language processing. There are tools like IBM Watson that can analyze text and, in theory, pull off a similar result. But Snyder says Textio’s results will be better because they are content-specific — like in the case of talent-acquisition documents.