categories

HOT TOPICS

Subscribe to our Feed

Optimizing Websites With Machine Learning: BloomReach CEO Raj De Datta (Part 4)

Posted on Sunday, Mar 11th 2012

Sramana: What was the result of your two year entrepreneur in residence experience? Did you emerge with a clarified approach for a new company?

Raj De Datta: When I was an entrepreneur in residence I started a cloud infrastructure company. I raised 500,000 dollars from the VC firm and then we experienced the 2008 meltdown. I looked at our business and felt that it did not make sense. It was going to be hugely capital intensive and it did not make sense anymore. I returned the money to the VCs.

I was just coming off of that experience and my co-founder, Ashutosh Garg, had just left Google. He had worked on search for four years. He worked on web search, product search, collaborative filtering and personalization. Around the time I met him he was working on building a search engine. I met him through a person who eventually became one of our investors, Justin Caldbeck. Just came from Bain Capital Ventures which was the original firm that provided the capital for BloomReach.

Justin and I had been talking and he told me that he wanted to back me in starting a company and he wanted me to talk to a lot of the portfolio companies that were owned by Bain. By then I had made up my mind that I was not going to build a consumer facing company. I was not going to build a company that was selling into IT. Justin introduced me to a lot of people who ran marketing at companies like Toys R Us, Michaels, and Dell. I spoke to those folks and the core problem that came out of those conversations was that they were spending huge amounts of money to get traffic to their websites. They wanted to know how to organically attract consumers to the content that they had in a scalable way.

When Justin and I met Ashu we knew he had been working on a search engine. He felt that at the end of the day a search engine is only as good as the content it knows about. If a user is searching for something and does not get great results it is more than likely that the search engine does not know about the best content that is out there. He was trying to build a better search engine. When we started talking we realized that the problem I was talking about, in terms of marketers having a hard time getting their content on the web, could be solved by the same solution to the problem he was addressing. That solution was a web relevance engine that will mine through the Internet, identify the best content and services out there, and deliver it in a well structured way for discovering search engines and front doors to the Internet. If we could structure that content well then it would offer a better experience to front end users. If users are able to find what they are looking for them millions of people are going to wind up visiting these websites.

Sramana: Essentially you were looking to create a layer between the content and the search engines.

Raj De Datta: Exactly. The fundamental observation that we made was what we call the digital discovery divide. There is proliferation of content out there. Millions of pieces of content are created all the time. Content is constantly being changed, yet every startup in Silicon Valley was creating yet another consumer Internet destination. If you play that story forward is that users only visit the same five or so websites every day. Thousands of startups are being created trying to become one of those five websites. The rest become a note on the Internet. They need to drive traffic.

This segment is part 4 in the series : Optimizing Websites With Machine Learning: BloomReach CEO Raj De Datta
1 2 3 4 5 6 7

Hacker News
() Comments

Featured Videos