Jesse Stanchak uses Ning's recent decision to focus entirely premium custom social networks to examine the viability of the freemium model. Jesse makes some solid points that Ning might have offered too good of a free product and that advocates of the freemium model should look to social gaming for ideas.
The key to a successful freemium service isn't just to get people hooked on the service but to offer premium services that are valued, not just differentiated from the free account level.
A free Ning network is pretty feature rich, which doesn't leave much room for premium services. The services offered (support, domain mapping, extra bandwidth, running your own ads, and removing Ning links) are nice, but likely don't compel enough free network managers to pay for them.
Ning would have been better off limiting the design customization and control capabilities at the free level and charging for advanced customization and full control. That way users could still create a customized social network but if they wanted it to be truly unique they'd have to open their wallets. If customization and control aren't what network managers value then Ning should have found what they do value.
The problem with going strictly to a premium service is that the customer will soon be presented with a formidable barrier to trial -- giving a credit card. Fewer people will start a network and those that do are more likely to abandon it if it doesn't take off right away. After all, who wants to pay for dormant social network?
Fremium is a solid business model. Companies just need to make sure they do it the right way. Focus on what customers value, see what competitors offer, and see if you can profitable deliver something people are willing to pay for.