What Do You Recommend?

While there’s no shortage of online sources that offer consumer reviews of places, products, services and entertainment, trusting a complete stranger’s opinion isn’t always a surefire way to guarantee making the best choice. Enter a new crop of recommendation services that offer more filtered, and more thoughtful, options to match each user’s unique taste.

Hunch: Hoping to build a business based on modern information seeking, Hunch culls data (known as a “taste profile”) about the articles, topics and people that users and their friends are “liking” and sharing on Facebook and Twitter, as well as from a series of questions designed to determine each user’s unique preferences. The service then uses “taste graphing” software to predict what other things those users might like. While other services offer consumers recommendations based on past purchases and specified tastes, Hunch is one of the first to offer information on things users might like even if their taste profile possesses no prior knowledge of their opinions in a given category. It can even be used to identify one’s next career.

Kick List: Launched in January, Kick List has yet to spur a media circus akin to that of Hunch. (Of course, that could have something to do with the fact that one of Hunch’s co-founders also started a little site called Flickr.) Yet, its “don’t trust strangers” mantra makes it a respectable entry among new filtered discovery engines. Anyone can become a member and set up a list based on a specific topic—anything from ‘best running shoes’ to ‘great party appetizers’. An invitation link to contribute to the list can be sent to friends who, upon becoming site members, can then add, comment and ‘kick’ things up the list. Think of it as a more readily navigable version of Facebook “likes.”

Google Places with Hotpot: Google’s effort to acquire Yelp may have backfired, but the search behemoth is becoming a player in the consumer review game. Because its new GPS-enabled Places app employs Hotpot, a personalized recommendation engine built on users’ ratings and those of their friends, it’s able to deliver results that are most relevant to each particular user rather than just the most popular ones. Say one is looking for the best Thai restaurant in the neighborhood. Taste of Thailand may be top-rated, but how’s one to know if those ratings were provided by users who actually know what constitutes good green curry or by those who just like the cheap prices? This app strives to make that kind of guesswork ancient history.