My colleague Brian Ascher in our Palo Alto office coined a term several months back called the “right-time web”. The concept, as he described it, was that with the massive increase in sharing of information through social and real-time media came the need for filters to help categorize that data and make it available on demand. It’s not that useful for me to know that a friend of mine just enjoyed an Americano Misto at Joe The Art of Coffee near Grand Central. But, when I am looking for great coffee places in mid-town, it’s really useful for me to know that. Similarly, when I am in the market for a digital camera, I would love to know who in my social and information networks has recently researched them and bought one. Their opinion is highly coveted by me at that time. What is really needed is a service to collect, organize, and make available all the data shared by my networks. Some think of this as social search. Brian called it the right-time web. And I think he is spot-on.

At the Twitter Chirp conference this week, I mentioned the need for it. Kara Swisher kindly wrote about it today and credited me with the concept, but it was really Brian’s concept and our team at Venrock has been riffing on it for some time. In any case, we have examined a bunch of companies in this space and would love to meet any others who are attacking the problem. Here is how Brian eloquently describes it:

Much has been written about the Real Time Web, and the hype grows louder every day.  There is no denying the power of Twitter and Facebook and the other real time social media sources to reshape the way we create and consume information, however “real time” is not for everyone.  Early adopters of these services relish in the ambient cloud of streaming information and the interpersonal “closeness” that comes from knowing the minute by minute moves of the people you are interested in.  Most of us however, don’t care what you just ate for lunch right now, but two months from now when I am in San Francisco looking for a lunch spot, I’d be very interested to know that you enjoyed and Tweeted about the charming café in Noe Valley from two months ago.  Likewise, hitting me with the triumphant news of your purchase of a new Prius is only modestly interesting to me while I am checking my phone while in line at the airport, but extremely interesting to me when I am proactively researching my own car purchase.  So, the point is that social media content is valuable because it comes from people you care about, but it is gold when it marries with your intent and interest in a topic at a time of your choosing.  In short, combining the “intent” of Search with the impact of a social filtered endorsement is opportunity of Google-sized proportions.  Just as marketers have found search to be an amazingly effective vehicle against which to drive performance advertising, if there were a search engine that provide socially filtered search results pulling from the corpus of social media content, that is the holy grail.

In our mind, it is a hairy problem perfectly made for data geeks. The real-time web is highly unstructured. We share links, photos, tweets, status updates, buying decisions, thumbs up movie and song recommendations, restaurant reviews, and the like. Most of this is broadcasted to the world and not well compiled. Facebook’s past stream is not really searchable and Twitter search lacks the filtering mechanisms of, say, Google Shopping. It also doesn’t allow me to filter by my social/info graph/friends.

So, for all of you data science folks who like the challenge of finding signal in noise, creating structure where none exists, and designing feedback mechanisms to see how users respond, we salute you and would love to help you solve this problem.

Thoughts? As always, thanks for listening and I’d love your input.