Reviews of products and services have become a currency of the web, helping consumers make more informed purchase decisions. If they have a serious fault, it is that recommender and reader may have wildly different tastes. Without understanding the perspective and lifestyle of a recommender, it can be challenging to identify the posts that are most relevant to you.
Hunch is a start-up that wants to address that particular issue by developing a deeper understanding of what makes you tick, so that you can have a better and more relevant experience online. They also seek to create a super ratings community, encompassing more product categories and playing more of a constant role in members’ lives.
All this understanding of you comes from connecting your FB and Twitter accounts to the service, along with information that you provide directly to the site. By collecting information on countless numbers of members, they are able to uncover what they call taste graphs, which relate to correlations between preferences for one type of thing and the likelihood of preference to other things. For example, consider this from a story on the company fromRWW:
…according to Hunch CEO and co-founder Chris Dixon, liberals do prefer arugula while conservatives opt for iceberg lettuce. The connection between lettuce preferences and political orientation is something that Hunch has uncovered through its taste graph and recommendation engine, something that Dixon describes as "the most sophisticated system ever built for predicting human preferences."
Members join the community and identify categories that are most meaningful to them. At the time of this writing, the site offered specific recommendations in the following categories:
- Food and Drink
- Movies and Entertainment
- Home and Garden
- Men’s Fashion
- Women’s Fashion
You and other members can read each other’s profiles and reviews and decided whom to follow. After you make ten ratings, the site will also identify individuals that have similar taste to you. As you consume information on the site, their engine is identifying your likely preferences, and updating its taste graph to reflect your actual preferences as they are revealed.
I wanted to see how good the engine was, so I joined and answered literally 685 questions relating to everything from political POVs to whether I am an over the top toilet roll person, or under the bottom.
I liked the experience that resulted. When you are transported into the community, you can turn on or off various product categories and consumer reviews that are recent and relevant to your likely preferences. Recognizing that my test was n=1, I found the content the site delivered to me engaging, and eclectically appealing.
As with all social communities, there is a hard core set of recommenders that are more engaged with the site and have delivered hundreds of recommendations. These people have also amassed thousands of followers, which doubtless also impacts the taste graph.
Brands can also apply to participate in the site. They currently make their money, however, on rev share with certain sites that sell goods the community members organically recommend. So, for example, reviews of books are linked to Amazon, as well as other review sites that members identify. Additionally, they have many partners who use their relationship with Hunch and its insights to make their on site or in app experiences more personalized and relevant to individuals.
A really fascinating community building a set of information that may have a profound impact on how we experience the web in the years ahead!