Fixing The Problem with Niche Communities and Sub-Groups
In the next release Community Server will support "Groups". I love the feature, but it could create a new problem for community site managers. How do you manage the creation and population of these sub-communities on your site?
I tend to agree with Sean and believe that Facebook does a terrible job of this. And it’s surprising to me because of the way they manage some of the groups functionality. For example:
Several Telligenti have submitted a request to create a private e-mail based group ala the @Microsoft.com group for folks at Telligent. We don’t get any reply to our request. No follow-up, explanation, or guidance on how to proceed. Facebook is clearly protective about allowing more .com groups on their site outside of the big guys.
The counter example to this controlled growth is found in their general groups functionality where anyone can create, join, and manage their own group. I think about the only value here, from my experience, is the often humerous group titles that show up in your news feed when a friend joins a group like "When I was your age Pluto was a planet". After that initial set-up nothing generally happens in the group. So what’s missing?
What’s missing is one part crticial mass and one part being drawn back into groups like you are drawn back to your facebook profile. Group specific events rarely show up in your news feed and that means you don’t tend to go back. That problem is easy to solve, but what about the critical mass question.
Lets say, for the sake of argument, that I’m a Red Sox fan. Lets search for a group to join. there are over 500 results in the group category of search results. The first page of results is a scattered mess with indistinguishable groups that range between 7 to 5000+ members. Which group should I join? Facebook, or any other site that allows groups needs to solve the problem and there are several ways to do it.
1. Allow groups to be sponsored and show them on top.
This may sound like selling out, but there is value to this. The Red Sox would probably pay Facebook handsomely to sponsor a group. Their official fan groups would be on top of the search results, fans could connect to the Red Sox more closely, and it would probably get critical mass to of people to create interesting discussions and content quickly.
2. Stop letting people create stupid groups or just delete them.
Or at least put some sort of gates around it. Your group inactive for weeks on end.. it gets deleted unless you do something. The "controlled growth" clearly worked well for facebook as the opened from college to college and from corp to corp… why not use that same request methodology here?
3. Strongly suggest merging groups or joining an alternative.
There is plenty of meta-data available to a social site like Facebook that would allow them to create good recommendations of alternative groups to join if you want to create a public group. The group creation page should show people existing groups that match your request at every step of they way. This could head off the creation of the the 500th Red Sox group. Once you do create your group how about comparing the properties of that group with other similar public groups and suggesting a group merger?
4. Improve Search Result Ranking.
I’m not sure how Facebook decided to rank the search results, but it’s clearly not by members or activity since there are several very active large groups on pages 3-10 of the search results of my Red Sox example. There are also dead groups on page 1. People are going to pick a group on the first page… make it the right set. It would be really simple to rank groups by size. People are drawn to a crowd and the problem might solve itself.
So my advice to community managers it to be careful about how you allocate your niche community groups. It can be a powerful feature for both public and private sets of people, but niche doesn’t have to mean zero activity.
So are you going to take your own advice and ensure these group revelevance and management features are implemented in Community Server?
Jonathan: I don’t think those changes would make them in, but I will be looking at getting them in soon afterwards. First get groups… then manage groups more effectively.
I don’t think, however, that just anyone in the community will be allowed to create groups be default so that will probably help significantly.
I couldn't agree with #3 more. I find the most annoying thing about sites like Facebook is their inability to at least limit options. I know at face value that sounds bad, but think about it: why do we need 50 Audi enthusiast groups that vary only by name…and the people in them. They're all after the same thing, but users don't typically think to search if their idea for a group exists before they create a group (and apparently ignore any potential "name already exists" errors). The end result is confusion for the users looking to join a group, perhaps even further ending up that users join no group because it's hard to tell which one is the "good one".
Agreed Heath. It really makes no sense to allow allt he seperate public groups since the users that joins group aren’t that likely to be limited to friends anyway.