Study looks at how to succeed with crowdsourcing

crowdsourcing-tap-500x360The last month or so has been something of a wakeup call for the crowdsourcing movement.  We’ve seen a study published claiming that 90% of crowdsourcing projects fail to get even menial levels of engagement.

What’s more, a second study found that even if an organization did attract a large number of participants, that often leads to rather bland outcomes.

With crowd based pioneers such as Quirky falling by the wayside, you sense it’s something of a reflection moment for the industry as the honeymoon period wanes and people have slightly higher expectations.

Taking a custom approach

A recent US study suggests that the best approach to these things might be to forget what other, higher profile organizations have done and to chart your own course.  What’s more, they even suggest that better ideas emerge when participants are not exposed to the submissions of others.

“Our research is among the first to explore how those platforms can enhance consumer performance by deviating from the standard structure,” the authors write. “Consumers who have a lot of knowledge about a particular area of an industry, for instance, generate better ideas when they are not immediately shown other consumers’ ideas.”

The importance of domain expertise

They were particularly keen to test out whether the domain expertise of the participant was important to the outcome of the process.  They tested things out via two experiments, one that looked for new ideas for the humble QR code, and the other that looked for ideas to improve consumer experience in various scenarios.

The outcomes of the two experiments showed that subject matter had a profound impact on the kind of ideas produced.  When members with limited domain expertise had access to that of more knowledgeable peers, there own contributions improved tremendously.

For those highly knowledgeable people however, the key seemed to be to break the ideation process down into subtasks.

“The best design is one that takes advantage of high-knowledge consumers, and that is a design that uses problem decomposition very explicitly and in which other consumers’ ideas are not shown,” the authors say.

Timing of disclosure

A study published late last year looked at the timing of any knowledge sharing.

First of all, the paper suggests that the benefit of a steady stream of information comes at the cost of diminished incentive for participants, thus potentially limiting the outcome of the project.

They also suggest that when disclosure occurs at the end of the process, it results in a lot more duplicated effort by participants, with a subsequent increase in collaborative work undertaken when information is disclosed throughout the project.  However, the flip side is that this independent effort also helps to bring about more innovative solutions.

When information was released throughout, effort from participants dropped by 26%, with roughly half as many submissions as when disclosure occurred at the end of the project.

All of which sounds bad, but when information was disclosed throughout, the participants built upon the work of others, so the quality of submissions then went up considerably, even when the quantity fell.


Leave a Reply

Your email address will not be published. Required fields are marked *