Unstructured Data Analysis coupled with Machine Learning is the most exciting area of business right now. In this interview, we discuss how Ulf and his team apply the process to Contract Analysis.
Sramana Mitra: Let’s start with introducing our audience to you. Tell us a bit about your background as well as Seal Software.
Ulf Zetterberg: I am the co-founder of Seal Software. I’ve been spending most of my time around content management. I have a passion for transformation of data to information, so it becomes useful and valuable inside a business process. I’m a serial entrepreneur. I’ve done a couple of startups, but I’ve also worked in large software companies. I was educationally good enough to work at a couple of large enterprise companies including EMC and OpenText.
Sramana Mitra: To work for those companies, you’ve been based in Silicon Valley?
Ulf Zetterberg: I move around the world. This is the second time I’m back in Silicon Valley. I lived in London, Barcelona, and Zurich.
Sramana Mitra: Where are you from originally?
Ulf Zetterberg: I’m Swedish by birth.
Sramana Mitra: Give us a little bit of the genesis of Seal Software. Where did you start the company? How did you get it off the ground?
Ulf Zetterberg: It was bootstrapped. My partner is Kevin Gidney, who is the CTO. Kevin and I worked together for a long time. The idea of Seal actually came at the back-end of the first e-discovery wave. As a popular content management firm, you’re always thinking of how to populate a new system with the legacy content. There were two problems. Customers didn’t always know where all the data content resides. When they know where the content is, it’s cleaned. There was a cleansing process.
We thought maybe the discovery could be the Holy Grail on the back-end of a search to help customers find relevant information. E-discovery led the way to Seal Software. There were some major problems with discovery. Discovery was a little bit better than enterprise search, but there’re still some holes in the stack. We focused on the problem that most of customers we worked with over the last years had, namely on how to take legacy information and consolidate information from multiple data sources into new applications. That was the idea that started Seal.
It’s a platform for unstructured data, but it also has pre-built rules and knowledge about content tags. In our case, it’s contracts initially. When we find a contract, we auto-extract and classify the information based on the knowledge we’ve built in the application. Then we can put the data in the right place. For contractors, there’re so many use cases. There’re a lot of touch points of contract, but we haven’t found a company that has one repository. That’s what I did with Seal Software.