Debating F2P Monetization

One of the things holding back the evolution of F2P gaming in the West is the understandable discomfort that many Western designers feel about the “aggressive” monetization strategies employed by Asian game developers. For the purposes of this post, I’m defining “aggressive” as the sale of items that impact gameplay and/or speed up a player’s progress, in addition to other, less controversial premium features like aesthetic items and account personalization.

To many developers, the idea of designing a game to be anything other than “fun” is heretical (they may also fear the possibility of offending sensitive players.) Consequently, they either ignore the F2P business model or attempt to create games with relatively tame revenue-generating systems; for example, focusing on the sale of items with aesthetic benefit only, or roping off a portion of the game and hoping enough players voluntarily pay for access.

The irony of these fears should not be lost on anyone who was designing games thirty years ago. Classic arcade titles were explicitly designed to eat quarters over brief, regular intervals, and people of all ages still put up with it. By comparison, modern F2P games are positively generous to players!

All this is why, up until the social game explosion, we heard of so few financially-successful F2P games in the West. The social gaming companies get a lot of credit for leveraging Facebook and for rediscovering the market potential of asynchronous gameplay, but they deserve equally as much credit for realizing that people in the West are not culturally predisposed to hating any game with an aggressive monetization model. As with everything in life, context matters.

Understanding the impact of conversion rates and ARPPU

Why is it worthwhile to at least consider the merits of designing a game with a more aggressive monetization model? It all comes down to conversion rates. The average Western F2P game is lucky to convert 5% of active users to paying users. At the low end, you get 1% conversion rates, which is where games like Farmville and Mafia Wars tend to sit. Some very rare games reach 20% or better, but to hit that level you generally need a fair bit of luck, an incredibly powerful brand and/or an intensely loyal niche audience. Bottom line: if you’re only going to convert 5% of your active users to paying users, you want to give those people every opportunity to pay you! Many of them will be delighted to do so if you handle the situation appropriately.

I’ve asked many F2P game developers to share their monthly ARPPU ( “average revenue per paying user”) statistics. Several have obliged in confidence, so I can’t share specific data points, but I can share averages. From my limited research, it seems that a game with a more aggressive monetization model and a loyal, niche userbase can hope to generate $50 per paying user per month, on average. (The term “average” is somewhat misleading — most users might pay $5 a month, while a small percentage of wealthier players might pay hundreds.) Obviously, these dollar figures will vary from game to game, depending on design, but they’re a useful generalization for the purposes of this post.

On the other hand, a F2P game that limits itself to flat subscription revenue and/or non-functional items is generally more likely to fall somewhere between $5 and $10 per paying user per month. You can expect the F2P equivalent of WoW (whatever that is) to do better than this, and you might expect a game that is largely focused on aesthetics to do better as well, but again, this is a useful generalization for most F2P games.

Different customers have different needs

Why is there such a big discrepancy between these types of F2P games? Basically: different customers have different needs. A game with a more diverse array of offerings is going to satisfy more people and earn more cash in the process, especially if it doesn’t arbitrarily cap the amount a loyal customer can pay. Some customers don’t have much spare time and are willing to pay for things that accelerate their progress. Some customers are mainly interested in making themselves or their surroundings more attractive. Some customers want anything that improves their social status. Etc. Customer XYZ might be willing to spend only $1 on aesthetic items, but $100 per month on functional items. Customer ABC may be the opposite. Every person is different.

There are other nuances to this issue. For example, the monetization strategies that convince a newly-active user to become a paying user may differ from the strategies that convince an old paying user to become an active payer once again. For example, imagine a game in which upgrading your avatar is an important (but costly) means of distinguishing yourself from newbs. A player might be willing to pay for the privilege of updating her avatar and distinguishing herself from newbs, but she only needs to do that once. How do you convince her to resume paying you? The answer, as before, comes down to having a diverse array of offerings that appeal to different kinds of players.

HappyFunTime: a fictional case study

To help put this in context, I’ve invented a fictional F2P game called “HappyFunTime”. You can play HappyFunTime forever without paying a dime (in other words, this is not one of those games that restricts most of its content to paying users.) HappyFunTime’s servers accommodate 2,500 active users per month. Each server costs $80 per month and eats approximately $120 in bandwidth per month. These fees consume a fixed percentage of HappyFunTime’s profit for every 2,500 active users, unlike payment processing fees (i.e. Facebook’s 30% rev share on credits) which are only incurred with paying users. There are other costs that scale with active users (for example, community moderation) but those costs don’t scale linearly, so I’m ignoring them for now.

If HappyFunTime combines subscriptions, aesthetic items, functional items, progress accelerators, etc, it can hope to generate $6,250 in revenue per server per month. That’s 125 paying users (5% of 2,500) paying $50 per month on average. Subtract 35% for all costs other than servers/bandwidth, and subtract $200 for server/bandwidth, and you get $3,863 in profit per server per month.

  • This is a niche game, so we’ll assume just 50k active users. That nets us a total monthly profit of $77,260. Not bad for a niche game that converts only 5% of its players to paying customers!
  • Server/bandwidth costs are eating just 3.2% of revenue in this scenario.

Now, for argument’s sake, let’s say that if HappyFunTime incorporates a flat subscription and non-functional items alone, it converts *twice* as many users to paying users. (In reality, I believe it would convert fewer players because it addresses fewer needs, but let’s run with this scenario.) HappyFunTime can now hope to generate $1,875 in revenue per server per month. That’s 250 paying users, paying $7.50 per month on average. Subtract 35% for all costs other than servers/bandwidth, and subtract $200 for server/bandwidth, and you get $1,019 in profit per server per month.

  • 50k active users nets us a total monthly profit of $20,380; approximately 1/4th of the profit in the previous scenario (or just 1/8th the profit with an equivalent conversion rate of 5%.)
  • Server/bandwidth costs are eating 10.7% of revenue in this scenario (or a whopping 21.4% with an equivalent conversion rate of 5%.)

In summary: because so few players actually pay anything for F2P games, the less aggressively you offer opportunities for paying users to support you, the less likely you are to be successful. And while it is possible to imagine a game that accomplishes this without selling functional items, progress accelerators, etc, that’s a hard feat to pull off.

Enough about money, what about ethics?

Some of you may still be thinking, “this still doesn’t seem ethical.” I can only respond to this by sharing how I feel. In my opinion, if the average person can enjoy playing a game for free, forever, without paying a dime, not only is the game’s design “ethical”, it’s practically charitable compared to the arcade games of the past. Or, for that matter, compared to $60 console games (given that I only have a few hours to play any given game, I frequently resent paying $60 for a bunch of content I neither need nor want.)

For that matter, I consider even the more aggressive monetization schemes in F2P games to be *far* preferable to the old TV model. Forcing me to watch 10 minutes of advertising for every 20 minutes of content feels abusive (if not akin to brainwashing.) I much prefer the opt-in monetization systems of F2P games.

Ultimately, ethical questions like this are highly subjective, and I neither expect nor wish to convince anyone of my opinion. This is how I feel about the work that I’m personally doing. Your mileage may vary.

26 Responses to Debating F2P Monetization

  1. On the question of ethics in this context, it may be worth drawing a distinction between two different contexts in which I’ve heard the word get used.

    1) Designing for monetization over fun: It’s this context you’ve largely used here, and I think it’s a non-issue. I don’t believe it’s unethical at all. Perhaps distasteful in the eyes of game design purists/traditionalists. It’s not *unethical* if there’s no dishonesty. Furthermore, I believe there’s a self-regulating feedback loop of sorts. If the design leans to aggressively in this direction, the “televangelism” becomes too apparent and the game ceases to be fun. To use your old-school arcade comparison: Games who’s 25-cent play lasted only 30 seconds, or who rewarded those that completed level 1 with a demand for more money to access level two, would have stood unplayed and failed in the market as a result.

    2) There’s another context in which ethics is referred to: that of misleading users and or advertisers (or anyone else in the value chain for that matter). The controversy around offers not too long ago that Michael Arrington brought to light is a case in point. If users were being duped into repeated cell-phone charges they weren’t aware of, or if they were being invited to game subscription offers repeatedly for acquisition bounties, these are clearly unethical.

    There are probably some gray areas in between the two. (For example, how small is the fine print on a Webkinz doll tag pointing out that the bundled subscription has an expiration date on it?). Still it’s worth drawing some boundaries between what are clearly ethical issues, versus just those offending the sensibilities of the gaming old guard.

  2. Thanks for this. I know we’re not entitled to the data points shared to you in confidence, but I’d really love to see more data than just the one salient value. Would you discuss the distribution of types of virtual goods that compose that $50 per paying-user/month? What % of that is progress accelerating, what % is functional, what % is cosmetic, ect?

    Thank you again.

  3. > I believe there’s a self-regulating feedback loop of sorts. If the design leans to
    > aggressively in this direction, the “televangelism” becomes too apparent and the
    > game ceases to be fun.

    Kim, I completely agree.

  4. > What % of that is progress accelerating, what % is functional, what % is cosmetic, ect?

    Patrick, that data is going to vary significantly from game to game — it depends entirely on the design. You can imagine a game that emphasizes aesthetics and personal expression having a different revenue makeup from a game that primarily emphasizes combat. I don’t have the breakdown for most of the games I investigated and I wouldn’t be comfortable attempting to “generically describe” the couple that I do have more detailed information about. Sorry about that; I’d love to provide more information but I need to respect the confidentiality of the developers who informed this post. :-/

  5. What you have shared is tremendously helpful!

  6. Chris 'Wombat' Crowell

    David, great article. I am sharing with our team. My major takeaway is rather a no brainer, the more player types you can satisfy with your game wares, the more successful you are likely to be. I think the idea that there are different kinds of gamer motivations is clearly understood by online developers, but it is still taking a long to sink in to the console gaming dev culture.

  7. Amber Venthus

    While you don’t have or can’t give explicit numbers like the percentage breakdowns, I was wondering if you could supply any information regarding the median ARPPU. Much like in real estate, I’ve found that the median values are often more telling than the averages which, as you noted, may be notoriously skewed.

  8. There are some inconsistencies in the post blog, as well as some examples that are very useful as they aren’t very relevant. For example, you say you resent playing $60 for a console game, but seem happy to pay $50/month for another type of game. You also say that you dislike playing console games because you only have a few hours to play, however if you extend that perception to your monetization schemes, a similar play would need to be monetized of $10/hour of game play (assuming for the sake of easy math that a few hours = 5).

    The other example you use is the coin play machines of the past. This isn’t a good example of peoples willingness to pay avatar upgrade fees or other ways of getting money from people. This is because the arcade games were largely skill based games. The better you were at them the longer you got to play for a quarter. I don’t think there is any equivalent in modern console or on-line games.

  9. I wasted hundreds if not thousands of my parents money one quarter at a time playing coin operated arcade games. (One time in Vegas at the Circus Circus I managed to spend more on arcade games in an hour than my parents did gambling.) However, I wouldn’t do so today and neither would my kids. In fact, I’m not aware of ANY arcades to take them to. That model got replaced by something better – console games and PC games.

    We’ve got an XBox, and XBox 360, a Wii and several PCs. What are my kids playing right now? Super Mario Brothers 64 on their Nintendo 64. So while many of those $60.00 games are a waste of money in terms of $/hour of enjoyment, others clearly aren’t.

    I think one thing that is lacking that would make these games more popular in the US is Internet Cafes. My (limited, possibly incorrect) understanding is that in some or most asian countries, most people don’t own a PC, they pay a small rate hourly to use a PC at an Internet Cafe. I can see how in that situation you might not mind paying a little extra for something inside a video game.

    In the US I think there may need to be a more careful balance: I can play WoW for $15/month and depending on my circumstances (available times to play, etc.) may be able to get the best stuff in the game. To me, a game for $0/month where I can _buy_ the best gear isn’t worth it. Much like Paintball and many other hobbies, the rich will have a great time and the poor, not so much. Now, if WoW cost $50/month, maybe it wouldn’t feel that way. Until then, no thanks “free to play” games.

  10. Back when Nexon was NXGames, there was a time when MapleStory wasn’t so cash shop dependant. Everything in the cash shop was for aesthetics only. Going with this idea worked for awhile, but then people realized that they’re basically paying for a bunch of pixels that don’t do anything, so they had to add 2x EXP potions and what not to actually start turning in a profit. Having an aesthetic only cash shop seems to fail in the long run.

  11. The main issues with monetization of non-aesthetic items which can actually make a player more powerful are primarily problematic in player vs player games. If all players have 100 health in a shooter, but for $10/mo you can have 200 health… in what way is that fair?

  12. I’ll preface this comment by admitting that I don’t believe games without player-versus-player combat to be worth playing. Clearly many people disagree (see the vast majority of casual games and MMOs ever made), but that’s their prerogative.

    Cash shops that offer ingame advantages in terms of skills, talents, or stats are a problem. Removing an element of skill in favor of “whoever spends more” will do nothing but cause the majority of your players to quit in frustration.

    Looking back at (though I shudder to mention it) Hellgate London, their monetization plan was well thought out likely would have worked well for them if the game wasn’t a steaming pile of fail (don’t blame the beta testers, we tried to fix it). Baseline content of the entire game for free, with paid extras like more storage, cosmetic items, and quicker access to future content.

    In any case, I fear that F2P games supported by item stores are the eventual future of multiplayer/MMO gaming. Interesting read.

  13. I don’t give a shit about what game companies want to do but whatever you do stick with it… if you want to do F2P with all the $$$ for time bullshit then do it up front. But don’t promise me one thing (that if I spend the time and effort I can earn something) and then a year or two deflate the value of my time by selling crap for $$$ so that you can cash in on the game I supported to get to this point.

  14. As a game developer and a long time player of massive multilayer games. My bottom line is versus server usage is.. OK lets assume my game garners the attention of 1000 active F2P player accounts in one month. Now these folks like my game and they play it allot. I at the end of the month have to pay for bandwidth etc. If 100 or so simply log on to play a short amount of time, they use bandwidth. I still pay for that bandwidth each month regardless if they pay for a feature or to unlock a quest.

    The greed part of this for me is maybe them paying for access to a feature or ascetics items then, yeah I could maybe maybe 30 dollars or even greater off them per month that a flat fee monthly subscription. but what if they don’t? what if all the time and work went into trying to find things that block them from being entertained and charge them a small fee for it,comes back to bite me and my company in the ass.

    It just seems F2P seems seems like pure greed, and designing content so I can charge them for a feature or quest or even access to dungeons or an instance is sadly a pure way to greed players out of their “Fun” Now granted,if my game is to much fun, then I run the risk of players abusing my F2P model.

    So for me and my company a flat fee each month will insure I can have solid players who pay the fee and full access to our game. Where I intend to make the extra income is by selling a name change or character transfer, But that about as far as I want to push them. I think the best idea for already established games is to reduce monthly fee to better match the current economy if players can’t get up 14.95 a month. not this F2P crap we herald in as the Saviour of dying MMOs.

  15. Been thinking about this for a while. Just read the comments. I see a bunch of people are anti F2P and I have been trying to find a way to defend the F2P model (I suppose I try to find a defense because I am currently designing a F2P game). I have been able to justify the F2P in-game purchases model only when thinking of it holistically.

    People who come to play a F2P game are not just a digital character in your game. They are real people each with a different level of access to money they can use to make in-game purcahses. This is a form of metagaming. The player is using their access to real world money (meta) to power up their in game character (gaming). There is nothing wrong with this. This model is actually more realistic than having a game world that is shut off from physical reality. F2P games with the ability to make in-game purchases merges the two worlds. So whether it is some kid with rich parents or a successful entrepreneur, it does not matter. The fact is that they have access to the funds. That is their lot in life and it is directly translated to the in game world.

    Granted this is game model is not meant for everyone such as more traditional game purists (those who rely on skill only to advance in the game) and those with little money (even though they will still play F2P games). Really, F2P games (w/ in-game purchases) and subscription based games (w/o in-game purchases) are two different beasts entirely. You can try to win the war on why this one is better than that one but it’s futile. There are gamers for each model. The choice for developers/designers on which model to build a game upon comes down to what your end goals are and what type of gamer you want to appeal to. That is it.

  16. The problem is that the subscription model is dead, or dying. WoW is the only western fantasy subscription MMO that’s turning a profit. Lotro went F2P and it started making a profit. EQ2 just went F2P and is making a profit again, and actually has players. Tabula Rasa closed because of not enough players. Warhammer Online is on it’s way out.

    Players don’t want to subscribe, because you know what that means? You are renting a game. Few people want to do that, and those that do are all playing WoW. Renting a game means not being able to play how you want, when you want. It means putting out money up front at the start of a month, and not knowing how much of that month will benefit you. It means cancelling the subscription when it’s not worth it any more, and it means avoiding the pain of paying for a new subscription, just to have to spend hours to ‘catch up’ with the other players.

    League of Legends is my favorite example. With real cash you can buy heroes, but they can also be bought with game money. It takes a while to earn a hero this way, but it can be done. Heroes don’t give an inherent advantage, as the free hero roster rotates, so eventually every hero can be played for free. Actual in-game power in the form of runes have to be purchased with game money, so there’s no buy-to-win situation. Lastly: unique skins for heroes cost real cash, so if you want to spend real cash then you can stand out. Players grow fond of their favourite heroes, and this gives them an opportunity to be different.

    In cash-shop games you sell customisation, convenience and time, not power or content. For example: the 2x EXP is so that people with money can level faster, but they pay for it. It doesn’t give more power, except for the fact the player will be a higher level in the same time. The freebies can just play more. If you charge for power or content, many players are going to cry unfair, but if the freebies can get it the hard way they might still cry unfair, but it’s unfounded.

    As for bandwidth costs, most ‘developers’ miss the point. You can’t have paying customers without a userbase to keep it supported. You CAN’T turn around and say “We only want that 5% that actually pay, and ignore the other 95%.” because those 5% *are* from the 95%. To only want to support the game for the ones that will pay is missing the point that the freebies are potential customers. That’s why there’s metrics like conversion rate and ARPPU. Think of it as supporting the entire population, and then work out how much the entire population is earning you. You are making a game to be fun, you are making a game to be played, and you are making it for profit. Those three things go hand in hand. Prevent players from playing your game by having an up-front fee and you’ll find yourself without players, and without profit.

Leave a Reply