My Hunch Is As Good As Jimmy Wales' Hunch

Wikipedia's Jimmy Wales

Wikipedia founder Jimmy Wales has joined the board of directors of Hunch, a New York-based startup that counts among its co-founders Chris Dixon and Caterina Fake. Hunch is one of the more important startups to emerge recently, yet it’s largely flown under the radar — much like Flickr did in the early days — despite its all-star founding team.

But back to Wales and his new director position. He outlines why he decided to join Hunch’s board in a blog post today:

I’ve always been intrigued by the potential intersection of community-based, user-generated web platforms and algorithmic, machine-based ones. Wikipedia and Wikia have proven to do a pretty darn good job with the former. Search engines clearly do a great job with the latter. But until recently I hadn’t seen a great example of how the two approaches could come together, co-exist and truly complement each other to form something greater than the sum of the parts – which I believe is the future of the web.’s combination of community-sourced content and algorithmically-driven smarts is forging a promising path towards this potential future.

Wales’ reasons are pretty similar to why Hunch is one of my favorite startups. At its very core, Hunch is a decision engine, but one that’s more sophisticated when compared to services such as Aardvark that offer answers to specific problems in a binary way. (Aardvark, by the way, is rumored to have received a $30 million acquisition offer from Google, which, if true, I think it should take, because as a service it will struggle to scale beyond a certain number of users.)

Hunch in a post on its own blog outlines its progress since its launch back in June:

  • Hunch users have created more than 5,500 decisions.
  • Hunch users have added more than 30,000 follow-up questions to those decisions.
  • Hunch users have contributed more than 50,000 decision outcomes.
  • Users have answered more than 28 million THAY (‘Teach Hunch About You’) questions.
  • Users have ‘played’ more than 6 million decisions on Hunch.
  • Hunch had more than 1 million unique visitors to the site in November.

Hunch (and companies like it) represent a smart way to deal with the increasing deluge of information on the web. Simply put, Hunch marries such information with your personal tastes to help you make choices. Using a decision tree structure, the site leads you through a series of questions, the answers to which help you uncover what decision you should ultimately make — whether it’s the next car you should drive or BBQ you should buy.

Caterina Fake

With the rise of the social web, “there’s an incredibly lush landscape of data that’s yet to be sorted,” Hunch Chief Product Officer Caterina Fake told me when I visited the company a few months ago. Caterina explained how in 2004, when Flickr was launched, many different trends converged to make the service a success: digital cameras had become commonplace, camera phones were on their way to entering the mainstream, broadband was becoming prevalent and blogging was hitting a new high.

She thinks a lot of different trends are coming together for Hunch as well, such as how people have gotten comfortable with the idea of sharing their personal information online, thanks to Facebook. (Remember the “25 Things About Me” craze on Facebook?). There is more trust in the collective knowledge shared on the web, thanks to Wikipedia. And most importantly, people are comfortable seeking answers from sites such as Mahalo and Yahoo Answers.

I see two big trends that favor Hunch: the pervasiveness of broadband and the explosion in the number of information-creation edge points, both of which are causing large amounts of information to flow into the web, challenging the search-based paradigm popularized by Google. For while Google can be the start of our quest for information, we need to augment it with services that bring the information to us.

In her recent post, All I Want for Christmas Is My Web, Stacey wished for a web that “comes to me on my phone, my navigation device, television or that fancy photo frame. It’s my web, delivering the information that it thinks I want, based on my preferences, friends and eventually location.” Well, Hunch is going to help her do that using a set of interesting algorithms including a “question selection algorithm.”

Indeed, as Fake explained, “We’re replacing calling friends and asking for her advice…we’re replacing human interaction, putting it online and making it easy to use instead of you going on to a search engine.” Hunch’s decision-based process makes it an efficient and targeted advertising platform. At the very least, the service can serve up ads for products that match the decisions people make.

For example, if someone uses Hunch to decide that a hybrid SUV should be their next car, the system could offer ads for car companies that sell hybrid SUVs. That’s what makes Hunch a lethal combination — one Jimmy Wales is glad he got behind.

Additional reporting by Jennifer Martinez.

Wales images courtesy of Wikipedia, Fake image courtesy of Duncan Davidson via Flickr.