We’ve gotten better as an industry about understanding communities and how to use them effectively but there’s still work to do. One issue I often see, even now, is confusing membership in a community with participation.
You might say that the two ideas are far apart, we know what each is and what it is not but there’s a lot of grey between them. Take participation for example. We commonly mean that people are participating in a community if they contribute something and since we’re dealing with the web, the contribution usually takes the form of some kind of post. It could be as little as one hundred forty characters on Twitter or something longer on a blog or more creative say posting photos on Facebook.
The person who is only a member is the contributor of most of the grey. Does that member actually log in to observe but not say or do anything or is the person a no-show. In reality it’s both but it’s hard to know which. The complication I see is when a community boasts its membership but not its participation rate.
In the last piece I discussed the need for knowing some demographic information about your members for the simple reason that you need to know the quality of the data on which you are making decisions. Thin participation in your target demographic can give the willies to the people signing the checks for a marketing program or new product development.
Recent research from a variety of sources illustrates how serious the challenge is for marketers. For instance, last year a study at Harvard Business School showed that ninety percent of the tweets on Twitter came from the ten percent most active members. Professor Mikolaj Jan Piskorski and MBA student Bill Heil conducted the study. Their data also showed that more than half of the 300,000 survey subjects tweeted only once every 74 days. Not exactly the participation levels most of us would hope for.
Another study by Pingdom.com (with a small sample) showed a small dominance of women over men participating on social sites. Women edged out men 53% to 47% in the average of all sites reviewed. But the ratio was as different as the site selected for study with Bebo garnering 64% women and Slashdot attracting over 80% men.
What to make of this? It’s all good, if you know going in what the participation is. If you want a predominantly male or female cohort to test an idea with you can find it, the real issue comes from not knowing.
Since so much of crowdsourcing and crowd wisdom depends on independence and diversity it’s simply important to control for it in your study. It may seem counter intuitive controlling in rather than controlling out in what amounts to a scientific experiment, but these are the new rules.