You May Like This...

Did you know that there's actually a technical term for the feature used by numerous social media sites that "suggests" what content you might like based on your past activity on the site?  Yeah, neither did I.  Collaborative filtering is being used by more and more websites to gauge a consumer's personal tastes and interests.  For instance, Amazon will suggest book titles to a customer within the same genre, or by similar authors via their "Customers who like this item also bought" feature.  Pandora maintains its livelihood by suggesting new music to a listener based on the artists they've been listening to for the past 3 or 4 songs.  Even clothing companies like Forever21 offer a feature on their website that suggests similar clothing styles, based on your past browsing (you don't even have to make a purchase in order for this feature to work.)

 Social media networks have also jumped on the collaborative filtering bandwagon.  Facebook has a tab on the homepage that shows the profiles of people that you aren't currently "friends" with but may know through a mutual friend.  Other Twitter users may be recommended to you because you're "following" their friends or vise-versa.  If you make a connection with someone from a particular company on LinkedIn, other employees at that company may be recommended to you. After watching a video on Youtube, a user can watch other videos that are similar in genre or contain the same tags or keywords as the video he or she just watched.

Pretty cool, no doubt.  But how exactly does it work? User-data profiling allows websites to monitor the preferences of a user over time, then adapt to cater to the likes and dislikes of that particular user.  The more interactive a person is on a website (such as posting a comment or sharing the website on their Facebook or Twitter feeds), the more accurate the profiling will be.  Since it's a technical, computer-generated tool, it's not error-proof.  There's not some human clone sitting behind your computer screen, watching your every move and picking out songs or outfits they think you'd like.  Still, it's a pretty innovative and interesting way for companies to give each customer a personalized experience.

Related articles: Spirit Magazine's "Just Like Me," Collaborative Filtering: Lifeblood of the Social Web,