What makes the difference between a Facebook Free2Play business model and lets say some MMORPG or browser game with an item shop is the social network component in the Facebook game. Basically the game company is not only interested in your money, they also would like you to get your friends to play. If that works, the game "goes viral" and reaches a lot more players than it would by just using traditional marketing. So far, so good. But does it work?
The problem, like so often in games, is one of balance. Facebook games are balanced around an assumption of how many friends you have and can bring to play with you. If you are outside that assumed range of friends, the model just fails. The problem is usually the use of progress blocks, where you can't advance any further unless you either ask your friends for help, or pay with a game currency you can buy for real money. If you have not enough friends, that makes the game far too expensive: I've been asked to pay up to $25 to complete a single quest. That is just silly, even for somebody who in another game might have paid $25 for a permanent mount. I do like having the option to pay for an advantage, but like most people I tend to react negatively to a paywall which basically tries to extort me with a "pay or stop playing" choice.
By making paying to play so expensive and annoying, Facebook games thus make the "social cost" of pestering your friends more appealing. That very quickly leads to players realizing that the person least likely to be bothered by a constant stream of gift requests is somebody already playing the same game. MMORPGs like Everquest started out with a social model in which guilds were there to play with your friends, and over time that social model degraded to guilds where you play with people who have the same goals and play intensity as you have, even if you don't actually like them. Facebook went through the same development much quicker. Every Facebook game forum has "add me" threads. My new Facebook account already has 67 friends, just by clicking on links in various "add me" threads like that.
Besides thus encouraging you to collect lots of fake friends, Facebook then also offers you to bot your interaction with those friends. I discovered a Facebook app called "Auto Collect Games Bonuses", which automatically searches through the feeds of your friends and clicks on all game requests for you, netting you all the bonuses you get for clicking on that sort of link. The app already has half a million monthly active users, and it is not the only bot doing this sort of activity. Apparently my 67 friends produced 655 of such links in the last 2 days, from 4 Facebook games (2 Zynga, 2 others). That's a lot of bonuses collected. This and the requests I can send them makes paying to play rather unnecessary, I can get around all those blocks by botted interaction with my fake friends.
I believe that this particular business model is not sustainable. Zynga already revealed that only 4% of their players are paying, and that percentage is decreasing. It is a downward spiral, in which players avoid paying by using the help of fake friends, and the developers react by making the game more expensive for the dwindling number of people still spending money, driving more and more people away from paying for the games and into the fake friends alternative. And of course by spreading the spam only to your fake friends and not your real ones, you don't even advertise the game to new customers.
There are a few games on Facebook that do better. For example in Dungeon Overlord you aren't required to have friends at all to advance. Inviting friends gives you a few percent bonus, with no hassle to you or your friend and no continued gift request spam. That is the kind of game which might encourage you to invite your *real* friends instead of searching out other already existing players and fake-friend those. But most Facebook games just copied the business model of the first successful games. And I believe that business model failed, and is going to come crashing down not too far in the future.