Netflix – The Greatest Crowdsourcing Example of All Times?

Netflix launched in 2006 and it was a huge challenge. Promising to be bigger and better than Cinematch, they set themselves what seemed like an impossible task. What they did to make their dream come true, however, was crowdsource. They asked anyone to come up with a piece of software that was able to predict movie preferences, and the winner would get $1 million.


Unsurprisingly, the crowdsource software that Netflix used for this became one of the top online hits within hours of the contest launching. Every day, new solutions were submitted and Netflix looked at which solution would be at least 10% better than Cinematch. This barrier proved hard to cross, and it took three years for someone to do it. That someone was BellKor’s Pragmatic Chaos, which was actually a team of seven experts from around the world.

Why Does it Matter?

The Netflix contest was popular for more than just the $1 prize. It was a pilot project that produced some amazing results. They made a database available that contained over 100 million ratings of around 18,000 movies, provided by 480,000 users. The results of this project could be applied across a range of industries, including politics, commerce, and science therefore.

This project really proved the effectiveness of crowdsourcing. Prior to the contest, Netflix had tried to solve the equation themselves but their successes were limited. They did not want to put all their trust in a single third party without having any guarantees. Crowdsourcing solved this issue.

Towards the end of the contest, various teams started to come together to join forces and find a solution, which even led to the creation of the eventual winner. This was inspiring for businesses all over the world, and particularly the scientific community. The solution included a blended mix of different models, each approaching the problem from different angles. This was why collaboration was so important, bringing together experts from each of those angles and models.

And best of all, the entire process was highly inclusive and innovative. It brought together some of the world’s best known and most respected academic researchers and scientists, but also included laymen who had never worked on virtual learning at all. Clearly, it identified talent from all over the world and it enabled these people to be recognized for their talents.

A Learning Curve

Since the Netflix challenge, a lot of other businesses have followed suit. Providing interesting prizes for good ideas is a great way to spend less money overall, while at the same time building up a database of other ideas, identifying and recognizing talent, and getting the actual solution to whatever the problem is that needed solving. What Netflix did was show the world that crowdsourcing works and that it is a very solid investment. Interestingly, in the end, they didn’t even implement the solution that BellKor came up with, but that is almost beside the point. They showed that the world is full of talent waiting to be discovered.

Leave a Reply