Serendipity - the new search Sunday, February 04, 2007
I confess that I was baffled when I both looking at the Aggregate Knowledge website and read their blurb and video on Demo 07. "Discovery. You know it when you see it." What does that mean in the context of a product?
However, a minute into the demo video I was immediately able to relate to the examples describes. Sometimes we stumble onto things by virtue of the actions of others - crowds outside a club indicate it's popular; we see the desert course ordered by a neighbouring table which looks appealing; someone is reading an article in a newspaper next to us on a train which catches our eye. Additionally, sometimes we search for stuff without being able to precisely articulate the "magic words" that will enable us to find it on Google - in which case you've no hope of locating it amongst the 2 million .
Aggregate Knowledge endeavours to utilise the wisdom of crowds [analysing their behaviour]. They acknowledge parallels with Amazon but are aiming to employ this for media and retail online businesses. They claim to be able to deploy it rapidly, citing a 16 day rollout for one large online store. They further claim that 20% of the Christmas turnover for this store was attributable to items "discovered" in the display/search panels they added.
It's almost like an aggregation of click stream data which presents you with the popular pathways that other people have followed based on where you are "now". This is less random than a service like StumbleUpon but more personalised than a simple popularity rating of all content on a website. It is also rendering information based upon "real-time" updates which is how it is able to service news sites, the content of which is constantly changing.
Most intriguingly, their revenue model appears to be entirely success based.
Labels: search
posted by John Wilson @ 9:46 PM Permanent Link
newsvine
reddit
