Towards Personalization: Strands Founder Francisco Martin (Part 4)

SM: How do you make your recommendation software portable to mass utilization?

FM: We have been working on something we call “Business Solutions”. The objective is to provide the recommender as a service for third parties. They can easily get recommendations integrated into their websites by utilizing our hosted platform. This is very cost effective.

SM: If I were an online retailer I could utilize your recommendation platform?

FM: Yes. Your cost is the cost of integrating our recommendation engine into your website by utilizing our platform. This is through an API like most other features, thus the cost is realistic. This is what I consider to be the second wave of the recommendation industry.
The first wave of recommender system occurred in 1998 and 1999. There were a lot of companies in the game back then. I know Amazon bought Net Perceptions which was a company founded by John Reidl out of the University of Minnesota, the inventor of collaborative filtering. Amazon was integrated Net Perceptions and other algorithms into their site.

SM: I had friends who were founders at Firefly, that was another recommendation engine right?.

FM: yes. It was bought by Microsoft and become the seed of Passport. One of the things Firefly was working on was the identification of users, so it made logical sense for Microsoft to utilize that technology for Passport.

Those companies constituted the first wave. The problem they faced came during integration. They needed to bring in a number of engineers, understand complex IT systems, and understand large product catalogues to successfully integrate their software into a clients system. That is a project that takes a full year to implement. It was a process which was very costly. I think that what we are seeing now is a second wave of recommendation engines which are API accessible. These engines can start providing recommendations via simple scripts. This is the way to go and there will be a bunch of companies competing in this market soon.

SM: Who else do you see competing in this space?

FM: Companies like Lumina and Criteo. Criteo is a French company founded three years ago which focused on video recommendations. About three or four months ago they got $12M of funding. Lumina is also pointing the right direction and they recieved $4M in funding three weeks ago. Telefónica was one of their investors. There are a handful of companies moving in that direction. The market is huge. There are a number of retailers out there who have very specific catalogues and do not have enough money to invest in sophisticated solutions. This is the way to reach those companies.

SM: Is retail your primary focus?

FM: Our focus is anyone with a big catalogue and who wants to figure out what users like and what should be promoted to users.

SM: Books and music should be a bit easier to correlate. Consumer electronics might be a bigger challenge.

FM: The great thing is you can not only correlate directly but you can correlate cross content. You can identify people who listen to this type of music and they are going to be more interested in a certain type of books or a certain place to eat dinner. There are a number of patterns in the way humans behave which we are unaware of because nobody has explicitly identified the trends yet.

With our approach over time we will know more and more about consumer behavior. The music consumers like, the videos they play, how many times they go to Starbucks, and if this is correlated to the music they buy or not. There are many people who do not believe humans are predictable but in reality we are very predictable.

This segment is part 4 in the series : Towards Personalization: Strands Founder Francisco Martin
1 2 3 4 5 6 7 8