To Succeed at Crowdsourcing, Forget the Crowd
The vast majority of crowdsourcing efforts fall flat, because companies don’t do enough to cultivate individual contributors.
When it comes to crowdsourcing and open innovation, we only ever seem to hear success stories. To be sure, the statistics can be impressive: Dell, for example, claims its IdeaStorm community website has fielded nearly 24,000 ideas since its 2007 launch, more than 550 of which were deemed good enough to act upon. Numbers like these would seem to indicate that there’s a world of innovative ideas out there for just about any company to tap into.
But crowdsourcing has two dirty little secrets. The first, as covered in my last piece for Knowledge, is that a flood of ideas does not necessarily equate to greater innovation potential. That’s because companies tend to favour more easily actionable - i.e. less innovative - ideas as contributions pile up.
As it turns out, however, that’s a problem available only to a lucky few. The second secret is that the majority - by which I mean 90 percent or more - will likely struggle to garner consistent feedback of any kind, let alone an overwhelming response. Looking at the performance of a broad cross-section of companies rather than just the successes, we can see how difficult it is to extract any value at all from crowdsourcing. To make it work, companies must concentrate not on the “crowd” but on developing reciprocal relationships with individuals.
The truth about crowdsourcing
For the research paper “Open to suggestions: How organisations elicit suggestions through proactive and reactive attention” (co-authored with Linus Dahlander of ESMT), we were able to lift the curtain on crowdsourcing thanks to an unusually rich dataset from a leading provider of online suggestion-capturing software. The data tracked a year’s worth of activity for 23,809 organisations that had installed the software, both suggestions received from external contributors and how organisations responded.
The results were sobering. Only the top one percent managed to generate an average of approximately one suggestion per day. The rest performed much worse. Organisations in the 90th percentile received fewer than 30 suggestions over the entire year. The bottom half on the performance scale - a group almost 12,000 strong - barely elicited any participation at all.
Paying attention pays off
It appears that companies’ baseline experience of crowdsourcing is not success but failure. Providing external contributors with the means to interact is far from a guarantee that they will. Perhaps blinded by the promise of innovation on the cheap, most organisations that build crowdsourcing initiatives lack a clear strategy for starting the flow of ideas.
We found, on the whole, that the better performers in our sample were much more generous with their time and attention than the empty-inbox majority. And they didn’t wait until the campaign had gotten going to jump in but were visibly engaged during the crucial early stages. They took it upon themselves to start the conversation as well as to sustain it with a consistent flow of attention.
Proactive and reactive
Specifically, we found that there are two kinds of attention that were important in this process: proactive and reactive. Proactive attention is when organisations become active contributors themselves and submit suggestions for community feedback. This shrinks the dividing line between organisational insiders and outsiders, acclimating external contributors to a back-and-forth that welcomes all pertinent ideas. It also gives the community a selective peek behind the scenes of the organisation’s decision-making, which may demystify it just enough that people feel their own contributions would be valued. This is most important in the early, most tentative stages of the campaign, when contributors are perhaps more comfortable responding to someone else’s ideas than posting their own.
Reactive attention is the responses organisations give to suggestions from external contributors. This could be as extensive as actually implementing a suggestion and announcing it publicly, or as simple as saying “thank you” for the suggestion. More attention paid is generally better for encouraging participation, but even a small response is better than none at all. Responding to suggestions proves the politeness of the organisation and makes contributors feel “heard”. It also may embolden lurkers to contribute by showing them what sorts of suggestions the organisation most values.
Responding to first-time contributors makes more of a difference than responding to established ones. Having no prior experience to go on, newcomers are particularly receptive to their treatment by the organisation. Paying special attention to them helps stave off the insularity that characterises so many user forums on the internet.
As with the proactive kind, reactive attention has the greatest impact early on in a crowdsourcing initiative. Unfortunately, most organisations do exactly the reverse: they are willing to invest only after the initiative reaches a more critical threshold.
Forget the crowd
With all this in mind, we can perhaps conclude that crowdsourcing is misleadingly named. When it does add value, that value comes not from the agglomerated wisdom of the community but from real relationships formed with individual contributors. Cultivating these relationships takes serious time and effort. Organisations that decide to venture into open innovation should first make sure they have the necessary resources ready to devote to the process right from the outset. Remember, the early stages are the most decisive.
However, our study suggests that even if everything is done right, many if not most attempts at crowdsourcing will wither and die, success stories notwithstanding.
Henning Piezunka is Assistant Professor of Entrepreneurship and Family Enterprise at INSEAD.
This article is republished courtesy of INSEAD Knowledge. Copyright INSEAD 2015.