I think Ning may be falling into a common trap. Just like many websites obsess over their visitor numbers, Ning obsesses over the number of networks created. This affects the site design: right on the front page, there's a huge "Create" widget that doesn't even check if something with the same name already exists. Try doing a search for say, "entrepreneurs." You'll get dozens of little networks, many of which have identical goals.

This duplication clearly hurts Ning: because of the network effect, merging two networks doesn't make it twice as useful -- it makes it four times as useful. I'm sure Ning knows this, but they aren't able to do anything about it because if they change the design, and the pace of network creation slows down, their investors might look at the chart and panic.

Ning is unique in their space, so right now they don't have much to worry about. But they have to be really careful about the fragmentation or else at some point they will have more networks than users and the useful activity will get drowned out in a sea of noise.

Some good points, but I'm not sure network duplication is really an issue here. If 100 people want to try to create an entrepreneur network and see if they can get a real user base going then that's fine. I doubt many users find networks to join by browsing through them on ning.com anyway, but instead probably come across them from elsewhere online. In that sense ning is somewhat similar to a hosted blogging platform. It obviously wouldn't make sense for blogger.com to try to merge similar blogs...

Someone should create a metanetwork that allows combining of multiple subnetworks and presents a coherent view of both.

well I think thats because Ning is not really a collection of networks with small sub-categories, but instead a collection of networks for actual people. So there may be more than one entrepreneur group who want to talk to each other

Neat. So how much money do they make?

I will be impressed with Ning's growth only when they start using the 'active user' metric. the spam/ghost town networks on Ning are a huge part of their "growth".

Take it with a grain of salt, but Compete shows pretty nice user growth as well. http://siteanalytics.compete.com/ning.com/?metric=uv

Now THAT, I'd believe in. 3.5 uniques a month seems reasonable and a good number to tout (though it's woeful compared to any other site with that level of funding).

But, given their 500,000 networks #, that means that they are getting.... 7 uniques per network per month? Assuming zero overlap?

And uniques are a bad metric, too. They have 1.5 million indexed pages ( http://www.google.com/search?q=site%3Aning.com ) - I'd wager 1/3 to 1/2 of their uniques are from long-tail search queries. And I'd further wager than 50%+ of those uniques BOUNCE (http://en.wikipedia.org/wiki/Bounce_Rate), so exactly what percentage of those 3.5m uniques actually do ANYTHING? Bounce rates of 30-50% for search engine traffic are pretty good, so...