Creative Commons License photo credit: davidChief

If you’ve ever visited a doctor about an ache in your shoulder you’ve probably been given the cliched-but-true advice “get a second opinion.” What if, instead of just getting two opinions you could get three? Eighteen? 48,132? That’s the principal behind crowdsourcing.

Crowdsourcing is – at a very basic level – throwing a task or request for information out into the public, and accumulating and selecting the best results. The theory is similar to “getting a second opinion.” You go see a second doctor because you want to see (a) if the doctors agree – which gives you a level of confidence that they are both right, or (b) if they disagree – in which case you’ve been given a level of confidence that at least one of them is wrong. If they disagree, you may need to seek a third opinion as a tiebreaker. If you take that and extend it to 10,000 doctors, you’re crowdsourcing.

Crowdsourcing – as a term – was first used by Jeff Howe in a June, 2006 Wired Magazine article (from Wikipedia). Most people are familiar with crowdsourcing through sites like Wikipedia, which uses tens of thousands of contributors and editors to create “encyclopedia” articles, on the assumption that errors and inaccuracies by the few will be corrected by the many. You don’t have to go far to find “offline” examples of crowdsourcing (Wikipedia even cites a few). When the World Trade Center towers were destroyed, the public was asked to vote on architects’ renderings for new construction on the site. American Idol is almost a form of crowdsourcing – instead of record executives plucking the next Britney or Hillary or Rhianna out of obscurity, AI relies on the assistance of the American voting public to select the Next Big Thing (with surprisingly mixed results – for every Kelly Clarkson there’s a Justin Guarini).

Crowdsourcing is evident daily in the New York public transportation system. Law enforcement is crowdsourcing surveillance, a tactic used by the US in World War II and (slightly more disturbingly) by the Soviet Union.

The Internet has made crowdsourcing simpler. Some interesting examples:

Each of these sites rely on crowdsourcing. Digg and Wikipedia, for example, effectively have thousands and thousands of people working diligently to provide, vet and publicize their content – for free. Manipulation of these services is attempted from time to time – Wikipedia quite famously has had bad publicity over paid crowdsourcing. Interestingly, the crowds usually supress efforts at manipulating content – but as the value of crowdsourced information becomes greater and greater, the efforts by third parties to manipulate the results will grow both more effective and more subtle. However, the results of sites like the National Journal’s Political Stock Exchange have been remarkably accurate over time – often more accurate than paid professional pollsters. Witness Senator Clinton’s comeback win in New Hampshire over Senator Obama, for example.

Since I learned of the term I’ve been fascinated with the idea. There is a book, The Wisdom of Crowds, that’s on my reading list that talks about this phenomenon, as well. The ideas of individuals are less and less reliable today, particularly coming from the Internet. You may still trust your neighbor when he tells you his doctor is the best in town. But can you trust BookLuvver439 on’s review of a book? I imagine you don’t. What you DO trust is looking at a book with 15 reviews that are all positive, and this trust grows with the number of reviews. Would you trust 486 positive reviews more than 3?

So how can you apply this concept to your personal life? First of all, rely on crowdsourcing wisely. If you read crowdsourced advice on losing weight, it’s what’s popular for the crowd – but it may not be applicable for you. Your tastes may be different (I was the only person in America who thought Titanic was boring, poorly acted and cheesy, I think). However, it can be a powerful tool. In personal finance blogs bits of advice like “invest in index funds” and “live frugally” and “buy a Wii” (just kidding) are so common that I would call them crowdsourced. Look for areas where consensus has evolved around the answer to a common question.

I want to find the next great crowdsourcing idea. I doubt I have the programming skills to make a website like Digg work, but all it takes is an idea. Could you crowdsource crowdsourcing ideas? The possibilities are endless.

When you look at my most popular posts on the upper right hand side of this blog – that’s crowdsourced. A lot of people stumbled my article on losing weight. Does it make it right for YOU? It might, although if you’re seeking to alter your diet or exercise I would always check with a health professional. A lot of people read those tips, liked them and thought they were good enough to recommend to the rest of their crowdsourcing “group” at StumbleUpon (my favorite crowdsourcing site). What I find really cool about it, though, is the fact that what other people choose as the most interesting articles are not the ones I would have chosen – that teaches me something about what works and doesn’t, and that’s always a lesson worth learning.

5 Replies to “crowdsourcing”

  1. The only problem is when the crowd runs off of a cliff (a la sub-prime, Bear Sterns, etc.).

    Question: Where does being contrarian fit into this?

  2. @No Debt Plan: True, and I guess that’s a critical distinction to be made: being contrarian is more about how you use the information that you’ve gathered, crowdsourced or otherwise, instead of the way in which it’s gathered. If I want to be contrarian about the stock market I’m still going to examine mass population attitudes (the run-up in gold is a good example of something exhibiting “dumb” crowd activity).

  3. You mention Hollywood Stock exchange, but what they’re doing at Media Predict is more interesting (

    They’ve gotten media companies to use a market system to actually put out content. So the crowd takes the role of editor, A&R rep, and more.

    I think market-based investment systems are the way to go for crowdsourcing. Mere voting (like at Digg) isn’t a good way to drive decisions — people have no incentive to give good information.

Comments are closed.