Critical mass. FB groups are a dumpster fire for topic discussions, but the fact that they're just there means loads of communities end up there anyway.
An interesting observation I took away from Google+ Communities engagement data (I had a full database of 8 million+ communities, most with membership counts, and the interactions on the most recent posts (up to 10) for each.
What I observed, loosely, was:
- There were far fewer very large communities than a strict log-log relation would have estimated, by a factor of roughly 3--10.
- Interactions per post seemed very loosely to peak at about 10,000 members. Given user attrition and various participation heuristics (e.g., the 90-9-1 rule), a 10,000 member community might be expected to have about 100 active members, and maybe 1,000 occasional participants, with the remainder nominally lurkers, and quite possibly entirely absent.
100 people is pretty close to the Dunbar Number, and would give an activity rate that would be sufficient to provide meaningful interactions but also a social cohesion.
I'd have had to look at the actual historyof these communities to see what trends and tendencies actually existed. I didn't, and the data are no longer available AFAIU. But what I did not, just from plotting activity against membership, was that there is such a thing as "too big".
It'd be interesting to see what the corresponding situation for FB is.
dredmorbius|4 years ago
What I observed, loosely, was:
- There were far fewer very large communities than a strict log-log relation would have estimated, by a factor of roughly 3--10.
- Interactions per post seemed very loosely to peak at about 10,000 members. Given user attrition and various participation heuristics (e.g., the 90-9-1 rule), a 10,000 member community might be expected to have about 100 active members, and maybe 1,000 occasional participants, with the remainder nominally lurkers, and quite possibly entirely absent.
100 people is pretty close to the Dunbar Number, and would give an activity rate that would be sufficient to provide meaningful interactions but also a social cohesion.
I'd have had to look at the actual historyof these communities to see what trends and tendencies actually existed. I didn't, and the data are no longer available AFAIU. But what I did not, just from plotting activity against membership, was that there is such a thing as "too big".
It'd be interesting to see what the corresponding situation for FB is.