If you're a good guy, you don't succeed, but if you're bad, everybody complains even though you're winning. Nice guys finish last?

The median request acceptance rate on Facebook has been hovering between 55-60% lately. That's largely due to gifting. I'd be surprised if anyone gets over 20-30% on normal invites, but with some good multivariate testing and an app that has broad appeal (such as Mafia Wars) that range is achievable.

If I had to wager it wouldn't be that you'd fail, it's that your perspective would shift a bit. The best thing about Facebook, rather than console or mobile platforms, is that you have the ability to continuously improve over time. It's not hit driven, it's more like web development. It took us a year and a handful of apps (some of which made decent money along the way) to get to where we feel we really got it right. We couldn't do that on the iPhone or Xbox platforms.

So you'd figure it out over time I'm sure, and your perspective would change. Mine certainly has. Stuff seems a lot less spammy when you realize that people are clicking accept 60% of the time.

Auto-tweeting is probably wrong, and was a mistake when we did it. We stopped that really fast once we realized that on Twitter, people don't expect it nearly so much as they do on FB. It's probably the right thing to do if you only care about growth, but we are more focused on engagement and keeping the customer happy over a long time, so in the end we felt that even ignoring the moral aspect it was the wrong thing to do.

I can't really give you much data on Twitter yet. We only started that game as an experiment and we haven't really put as much time into it as we should. Our game on Facebook was growing like gangbusters and we had to scramble to keep up. We're probably going to put some more work into it soon, so I might be able to give you some more insights, though unlike Facebook I'll never get to see how we stack up against the median.

I'd love to see response-rate data… i.e. if they send out a million in a
day, how many clicks happen? The harder data to get at is “how many people
are annoyed and how annoyed are they?”. Obviously they work. I wouldn't
classify them quite as low as viagra spam… A better email analogy might
be the “newsletters” I get signed up for without my permission. Most of
them are intelligent and informative– and I Imagine they are useful for
some of the recipients. I'm sure those perform well, too. I imagine they
are well-received for other active players– I assume your wife plays Farm
Town.

But that's the problem. Spam started that way– well written and
informative (if unsolicited) email that was trying to be targeted. If it
keeps working, people are going to keep doing it. It's going to get worse,
because it's a really effective medium. Response rates will go down and
aggregate annoyance levels will go up. I don't see this getting better for
folks like me!

The system is clearly rigged for this– I totally agree. Obviously, there
are degrees of high road and I probably wouldn't take the HIGHEST! Some
people who take the highest roads still kick ass. There are plenty of games
that have had great success with zero built-in virality. Good ol' fashioned
word of mouth and marketing (from MUDs to World of Warcraft to World of
Goo). I think this is a more hit-driven path and if I ever set foot on it,
I'm sure I'd bomb heroically like most starry-eyed game makers do! I'd do
it for the love of the game and not necessarily to chase the best path the
wealth and prosperity.

Anyhoo, thanks for the great points and congrats on your success with Blue
Frog!

I currently have 1 friend playing Foursquare, and its almost annoying enough with 1 to think about blocking. If I'd done that already, I'd have missed reading about his recent trip to Machu Pichu – and that would have been sad.

1. One man's spam is another man's ham. The line isn't as clearly-drawn as you indicate. I hate the Farm Town gift requests as much as anyone, but my wife loves them, and they've driven the median request acceptance rate up from ~20% to ~60% on the platform. The fact that these tactics work (and it's way higher than 1% response rate) indicates that a sizable number of people like them.

2. The platforms (especially Facebook) are designed in such a way that if you don't spam, you get choked off and die. Seriously. You cannot make a popular game on FB without gifts anymore. The platform is fundamentally broken in that respect. I'm not sure if you can make a popular social game on Twitter at all, but if you can, you probably have to cross that line.

3. Most of the popular web services we know and love were built through such spamminess. Facebook got big by asking you for your email login and password and then inviting everyone. Is that not spam? Sure you agreed to send those (just as you agreed to send the pigs on farm town) but the receiver never asked to get that stuff.

“I think I’d just focus on making really fun games, making it MORE fun if people invited friends, and giving them the tools to tell the world should they want to.”

You'll still be accused of spamming. The person annoyed at receiving 20 Farm Town gift requests per day doesn't really care how much the sender liked the game.

The problem is not that game developers want to spam, it's that the system is designed in such a way that we have to do it (in which case we can make a fortune) or we fail (in which case we lose one). If you want to fix the problem, fix the platform, which would be easier if any of the platforms had any real developer outreach. Unless you're Zynga it's nearly impossible to just get anyone with any clout to respond to an email.

If you insist on “taking the high road” then I'm glad you're in time management.