Power To The People-Powered Recommendation Engines

9 Dec

A few years ago, Netflix offered a $1 million prize to anybody who could create a better movie recommendation algorithm. To collect the grand prize, the algorithm had to beat Netflix’s by 10%. To date, no one’s collected the cool million. People-powered search engines have an equally dismal track record: when’s the last time you used anything other than Google?1 But people-powered movie recommendation engines have more promise. Taste in movies is far more intimate than general searches. Music recommendation engines have split the difference: Pandora relies mostly on algorithms, while last.fm tries to integrate a more social aspect, which makes it far more fun to use.

So it’s not surprising to see Clerk Dogs enter the field. Skipping the Netflix Prize goldrush entirely (and started by a former video-store entrepreneur, not surprisingly), Clerk Dogs is an explicitly human-powered movie recommendation engine. Their business model’s straightforward enough; Amazon referrals should make it fairly profitable if traffic picks up. For now, it’s in private beta (invites are still available), and a LOT of rough edges need to be smoothed out, but if this ever gets off the ground, it could be massive. While Flixster might have cornered the market for Facebook movie widgets, no one’s made a serious play to be the last.fm of the movie world. Given the size of the check last.fm cashed when they sold to CBS, it looks like there’s money to be made, and the field’s still wide open.

Clerk Dogs

1 Seriously, Mahalo? Really?

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: