Paying for Peering

paying-for-peeringCollective intelligence is touted as a rich resource that companies need to tap into in order to maximize profitability, but is it something they are willing to pay for? Not so far. Until the Fortune 500 start stumping up some serious cash to invest in crowdsourcing it will remain more of a buzzword than a phenomenon. Of course, we can point to Wikipedia, Goldcorp, Linux and others, but tapping the crowd as a strategy is far from widespread as a for-profit strategy despite having a few examples that make for interesting reading.

In the 2006 book Wikinomics, the authors outlined their belief that peering is optimal when three conditions are present:

  1. Object of production is information or culture, thus keeping cost of production low for contributors
  2. Tasks can be divided into small portions, thus making it easy to contribute
  3. “The costs of integrating those pieces into a finished end product, including the leadership and quality-control mechanisms, must be low”

I don’t agree with the third point. If we are going to move on from encyclopedias and logo competitions, then it is my opinion that the technologies designed to leverage all the benefits of crowdsourcing warrant sizable investment. If peer production of goods and services has half the potential that its advocates claim, then this investment will be well worth it.

It is time for companies to start paying for peering, and invest in a resource that is as untapped as Goldcorp’s assets used to be. I’m looking forward to 2010: with more and more companies, such as Chardoix, builidng platforms to help access the crowds we might finally get to see what we can do.

Share this post with the crowds?
  • Sphinn
  • Twitter
  • StumbleUpon
  • Facebook
  • Digg
  • Google Bookmarks
  • LinkedIn
  • MySpace
  • Technorati
  • Yahoo! Buzz
  • email
  • RSS
  • Reddit

Tags: , , , ,

blog comments powered by Disqus