Skip to content

Emergent dynamics in Facebook games

March 6, 2009

A few days ago, I talked about the primary business model of Facebook games: give us money, visit our sponsors so they can give us money, or propagate the game to your friends so maybe they will do one of the above. Here’s two examples of why this model is broken.

Attack of the Clones

In Warbook, the primary mechanic revolves around how to allocate your resources: towards resource generation, attack, or defense. You’re encouraged to team up with friends with the ability to send resources to other players, and two players who coordinate and assist one another are going to be more powerful than if they just played separately.

However, what’s even more efficient is if a single player controls multiple characters: several that are pure resource generation with no attack or defense at all, that funnel all their resources to a main military character. This allows the main character to field a much larger army than they would otherwise be able to support.

This is not difficult to do. You can create free email accounts at a variety of places, and create a free Facebook account for each of those, and then add these alter egos to your friends list.

So, the game actually rewards you for creating “alt” characters, over and above inviting your friends. So much for propagation.

About the only thing the developer could do is to make “alts” against their terms of service (which is what Warbook does). And then they have to devote a substantial ongoing support cost to the game playing whack-a-mole with players who abuse the service, get kicked off, then create more Facebook accounts and come right back. Or else don’t bother, and then the policy has no teeth and is blithely ignored anyway.

Friend your friends.

In the game Elven Blood, progression is blocked at several points until you have enough people in your “adventuring party.” You get more party members by successfully inviting friends into the game, of course. You may add your friends to your party if they already play, though, so this is not guaranteed propagation for the game — you might be “propagating” to someone who already plays.

The other day, I found out this can go in reverse. I received a friend request on Facebook from someone who plays this game and is looking for new party members. (In order to add someone to your party in Elven Blood, they have to be on your Facebook friends list, so he couldn’t just add me in the game without Facebook-friending me first.)

So, here is a case where not only has the propagation method failed (this player is friending people in the game, rather than propagating the game to their friends who don’t play), but it has the interesting social side effect of introducing two people who never would have met otherwise. Fun for me, not so great for the developers.

Don’t worry, we’ll do it for you

Recently in the game Mob Wars, the developers added a new feature: it will automatically search your Facebook-friends list and tell you which of your friends are already playing, so that you can “friend” them in the game too.

I assume the developers added this feature from popular demand, because players would incessantly ask for this to be made easy for them. The down side is that it gives people large groups without having to propagate the game to someone new. The developers get absolutely no benefit from this feature, and have (unfortunately) just destroyed half of their revenue model in the process.


There are too many easy ways to abuse a propagation mechanic. Facebook games have to limit their rewards for inviting friends, and focus more on monetary rewards (paying the developer directly, or visiting sponsor sites). Propagation does not have the value that a game developer would like it to.

Finding alternate monetization methods (without relying on propagation) should not be that hard. Non-Facebook games like Kart Racer and Maplestory somehow manage to do well with the free-to-play microtransactions-for-premium-content model without any built-in requirements to invite friends.

– Ian Schreiber

4 Comments leave one →
  1. Kelly permalink
    March 9, 2009 12:28 am

    The “business model” for Zynga games like Mob Wars, Fashion Wars, Pirates!, etc. was to gather inflated player number stats so they could secure obscene funding rounds for their company. ;p

    I don’t think these games are as reliant on what you termed “propogation mechanics” for revenue as you might think. Also, for many facebook games the point is to deepen the active player’s interest (increasing their money and time spent with the brand/game) not to rack up legions of “invited players” who have zero interest in the brand/game and will not invest time or money playing.

    Your take on this seems to presume that there is some economic value for a game to have tons of invited players who accepted a friend’s invite only as a favor to the friend. These reluctant, inactive “propogated” players are not valuable at all to companies like Zynga (outside of plumping their user numbers, which they hardly need at this point now that they are substantially funded).

    I just designed a social game where the client wanted release on FB Apps and the upcoming J2Play platforms (myspace, FB, Bebo, hi5, etc.). The main goal was to engage players who were genuinely interested and active with fashion online and to deepen the player’s time spent with the game by letting them gain entertainment value from inviting and sharing the game with interested friends (friends who would be intrigued on the game without being panhandled for halfass participation by their player friends).

    Games exist to entertain people so I don’t understand why you would say that “developers get absolutely no benefit from this feature, and have (unfortunately) just destroyed half their revenue model in the process” when the key thing any publisher wants is happy active players. A feature that makes players happier in your game is what you need to achieve viral word of mouth so that other people, valuable potential players who genuinely are interested in your game, will come to the title organically.

    Also, note that the FB Apps platform is very clear now that you can NOT reward players directly for inviting friends (with points, etc.) so it is well recognized now that spamming invites is not a valid gameplapy mechanic and that the platform itself will suffer if virality mechanics are the only means a game has of attracting new “players” (most of whom don’t stay to play).

    Anyhow, I think the “propogation mechanic” term is more commonly called “viral feature” and is quite common in web app design now. It’s in no way a business model and I have not seen any game relying on it as one, especially the Zynga type games on FB.

    Their primary driver for generating revenue from players is the players LACK of willing friends to participate fully in the game. Also the lack of resources when you are trying to move up the early levels of the game. There is so little to do gameplay-wise that players are compelled to earn the “favor points” (sponsored deals, etc.) just to keep the game moving along. The average player is annoyed trying to figure out how they will gain 20 players in their posse without spamming their nongaming friends and family. Especially in the first 5-20 levels of the game, there is a lot of attraction in “just doing one or two” of the sponsor offers so you can get a leg up quickly.

    This is the equivalent to getting into WOW and having no quests, no energy or health (cuz replenishment is on a 60 minute timer) but there’s a murloc who’ll give you full health and a Newbie Sword of Desperation if you sign up for a Netflix account. The player is eager to get on with things, impatient that there is no gameplay mechanic to help them progress or grind ahead, so this makes the lame sponsored offer palatable when it would normally be ignored if it were just a traditional banner ad in the sidebar of your FB profile. The game gives the sponsor a receptive audience. It creates a motivation point that has players seriously considering whether they will take that Netflix trial account when really they didn’t want one OR wanted one but had some reservation about it.

    That’s the business model there, and it works surprisingly well compared to traditional banner ads where your sponsor is lucky to get a click-thru rate above the standard .5%. These sponsored offers may not get a ton of click-thrus but I bet they get insane conversion ratios because by the time the player is clicking they are committed to giving this a go so they can get back into the game. 🙂

  2. March 9, 2009 8:24 am

    Wow, Kelly. Great to hear from someone in the biz! Unfortunately, my analysis has no insider information and is based purely on my interpretation of the mechanics.

    You may say that Mob Wars doesn’t care about player propagation. But:
    * One of the main determiners of your power level is how many “mob members” you have;
    * You can get a new member by successfully inviting a friend to play the game;
    * Or, you can “buy” two new members by completing offers / paying cash. It costs 10 Godfather Points for this, and you can purchase these at approx. 25 cents per point (or less if you buy in bulk).
    * Therefore, 1 friend invite = 5 Godfather Points = $1.25. Zynga may not need me to invite my friends, but from their business model it would appear to have cash value regardless.

    I think you’re right that propagation as the only mechanic fails miserably (as we all found out during the “summer of biting chumps”). But even with modern games it does still seem to be a mechanic that doesn’t go away, and is integral to the gameplay.

    Other games are worse. Elven Blood literally locks you out of content, with no way in, unless you either invite friends or pay cash. It’s hard for me to not call that a primary business model. And that’s one of the “better” games out there, judging by player base.

  3. Kelly permalink
    March 9, 2009 7:13 pm

    Good points. I have not played Elven Blood (will check it out now) but had to research similar social rpg games (particularly the Zynga family of “wars” style games). I’m sure some of them do the lock out more clumsily than others and make it seem like the propogation/viral is the only thing the game makers care about. No idea why you would block people *completely* from advancing unless you are trying to recreate the old shareware model of x levels free, pay for the rest.

    And for a smaller, less-funded company, I really think growing your player numbers IS the key metric (to attract more funding, clients, etc. it’s just not impressive to say “XXX,XXX players” anymore, you almost need to have millions of players when you’re on these social nets). Once you gain respectable traction then I think the focus shifts to revenue per active player (or just plain revenue per player though that would probably not be the best metric when you are pushing players to invite inactive players just to increment their group/posse/mob type stat).

    My theory on the real purpose of friend invites in a sophisticated social net game by a company that cares about revenue per active player (and player satisfaction) more than, say, pumping up their total player volume/exposure is that it creates scarcity in a key game resource because most players will not want to violate subtle social norms (spamming non-player friends) just to get ahead in the game.

    When something is on a timer, even an impatient person will grudgingly admit they can do something else (play another game!) while the problem solves itself and their timer resources replenish. But the size of your group does not grow passively like that. There are millions of people you *could* invite (and vice versa) but there are social norms that make us feel that there is really only a small subset of people we CAN invite without squandering other more important types of social capital. Power players get around this by friending total strangers who actually play the games (and thus theoretically won’t be judgmental or put off by the invite) though that sometimes feels very “un-Facebook” and more like the MySpace friend-anything-that-posts era.

    By creating this pivotal metric that is awkward to grow naturally it creates a really strong incentive for players to use those sponsored offers which offer huge blocks of favor points.

    I really wish there was a white paper or something aggregating publisher experience with these revenue models. I would love to know WHAT most players spend their “favor points” cash resource on. I am betting it’s mostly extra “bought” friends to add to your groups and cash in the lower levels (though in all the Zynga games it only takes 9 or fewer levels to become a cash machine, like where you log back in after a week of being busy and find $13 million built up).

    Interesting things to check out if you’re pondering some alt rev streams (missed opportunities I’ve noticed):
    -Zynga’s Vampires game now has a slot machine kind of feature where a spooky kid spins a demonic wheel of fortune and you get a random group buff (applies to your friends too) or curse. Paying favor points along with your spin ensures the payoff will be something helpful. This is totally underused- letting people engage in higher stakes (more exciting!) challenges but then offering a way they can micro-pay for “insurance” to improve their odds of success.
    -Zynga’s Dragon Wars game now has flash 3D combat rounds mixed in as optional quests with the regular ones. These take longer to play for the same points as the easier ones but could be improved to offer superior entertainment for players who like point-click MMO combat. However the controls scheme is freakishly bad (you use your mouse to move, aim/attack, AND select an offensive skill via awkward mouse gestures prior to every attack click). IMO, if these could be made multiplayer then you could form PUGs for live “raids” and add a really amazing social dynamic to these little games. It feels like an area that could be an upsell or encourage more of the high-stakes risk mitigation incentive thing (for example, we all know it takes 15 minutes or more to run this raid and we really want the best drops in there so everyone in the group pitches in enough favor points so we can run the Epic version which has higher probability of high end drops)
    -Zynga’s Fashion Wars and Mafia Wars are set in contemporary context where brand advertising would work well. Having a Gucci sponsored Gucci bag with a live affiliate link to buy the bag from the game would be a natural, but there is also a greater chance people will buy the virtual version of the item if it has real world brand. It would also be easy (and good viral bait) to let players gift the virtual brand items to other friends (players and non players).

    For social network games the main problem with micro-payments is getting someone to care enough to go through a transaction for any of these games. If there was an easy way to buy premium content on the platform (a FB wallet) then it might be more viable but right now I think people’s emotional investment is not quite high enough to make them break out a card for all of these little games. Maple Story has a lot more to offer and stronger social status elements (see/be seen in real time chat environments). Micropayments are a hard sell for a little game like Bar Brawl.

    I also think there is a HUGE advantage to just using FB Apps as a pure advertising platform for a full version of something. I’d love to see a good flash game like Auditorium produce a FB version (maybe you only get the act 1 mechanics but with random puzzle generation/many variants). Then the upsell would just be the full version of the game, but it would have to be something really compelling and unique. Not many FB games have those qualities.


  1. [Link] Modelo de Negócios no Facebook « Maracatu Studios

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: