We use the freemium model with our time tracking/pm/billing app Paymo succesfully. I don't think the model is finished, it really has great potential BUT it's not a magical formula that works everywhere.
The way we do it with Paymo is that we offer basic functionality for free to people who can't afford the software at that point in time. (eg: someone who just starting freelancing). In time some of these users convert to paying customers.
The major problem developers face at the moment is the huge availability of "free" - some users just assume that they are entitled to free software no matter what. You really need to provide real value and have your users perceive this value in order to make money.
At the moment we have a bubble forming with huge amounts of cash poured into businesses that will never ever earn any real cash and i think it will pop pretty soon. This creates a lot of problems both short and long term which we'll have to deal with.
I'd say that there are quite a few business models where you don't need to provide real value, or even have the people you are getting money off to perceive any real value and still make money. And some of these business models are even legal.
The console and pc video game industry makes plenty of money just by charging for their software. I think App developers need to stop this race to the bottom thats killing everyone. I guess its fueled by this idea that on the internet the number of users is paramount to everything. Which I think is a silicon valley culture thing that seems like its going to cause or has caused a bubble.
I'm not necessarily disagreeing with you, but you mention the gaming industry; I thought I'd point out the biggest new development there is giving away the games for free and charging for items, aka. F2P (Valve/TF2 is one example. They increased their revenue 12x[1]).
Anecdotal but from what I recall, the freemium model was an import from Korean MMOs which then got cross pollinated to other systems.
As I remember it Western MMOs/games focused on the subscription model, aside from WoW were suffering but were aware of the Free 2 play model in Asia.
That got imported over and soon after it was shown to be viable, the idea propogated through the system
So the original model is pretty old, and that is also the first place I heard about a freemium model before it began being heard in non-gaming enterprises.
Valve is not a good example for anything. They do things other companies can't, or won't, and at any rate, don't. They are the exception that proves the rule, in most things. In the TF2 case, they went to that model as an experiment, after TF2 had already been paid software for a long time and had made a lot of money for Valve. So, you don't see much of the annoying 'monetization' bullshit that you see with other titles, because there wasn't any pressure to put it in. Instead, they sold things that people wanted without impacting core gameplay which is something that no other freemium developer seems able to pull off.
I haven't encountered other examples of freemium games of that type, that weren't utter garbage.
Somehow people seem to have forgot along the way that value != money. I am a huge fan of people doing things just because they are useful without worrying about how they are going to try and monetise them, but just because something is useful and used by masses of people doesn't mean that the network of people that use it is easily monetisable, especially after the fact.
It seems that people just looked at the number of users and assumed that because a company can get millions of users, that the people working there must be bright enough to work out how to get loads of money out of them eventually, while forgetting that you don't need to be that smart to get millions of users when you are providing a useful service for free.
1) Ridiculously Useful
2) No Comparable Competition
3) Has A Tie-In Factor That Makes Any Power User EVENTUALLY Pay (and be more than happy to).