Microtransactions and other drugs
How much does a dollar cost? It depends. An essay about Minecraft server monetization!
Update: Apr. 17, 2020
As of April 17th, 2020, Mojang has released a new edition of their Commercial Usage Guidelines (with an accompanying blog post). The new guidelines have some significant changes - accordingly, some of the commentary on this post is now outdated. In the near future, I will release a “version 2.0” as a new post (and link to it here) with new commentary on the updated policies. In the meantime, please read the new Guidelines and blog post from Mojang.
Hello and welcome to the first of what will hopefully be a series of essays and commentary about game design - especially as it relates to running Minecraft servers! I’ve been writing this post for about three months now, and I’ve spent a lot of effort on writing commentary that is both broad and deep. My goal in writing this post was for it to become a useful reference point for server owners both new and old when thinking about how to monetize their servers in meaningful, EULA-compliant ways. I had a blast writing it, and I hope you find it helpful.
Introductions and disclaimers
Monetizing games is hard, to be sure. But any developer - especially a small-scale developer - has at least some freedom of choice about how to shape their game’s monetization strategy. This post is a discussion of some popular strategies.
Today I’m going to talk mostly about monetizing Minecraft servers. It’s what I’m most familiar with on a large scale¹ - that is, across multiple different servers - and it provides a stable platform by which server operators and developers can make their own economic and design decisions. Minecraft server owners are designers, even if they don’t realize it! Many owners insist that they’re just copying the systems they see on other servers, but I find this excuse to be unsatisfying. Every server owner is responsible for how they build and fund their server.
By way of background, Mojang (the company that develops Minecraft) has published a EULA and accompanying Commercial Usage Guidelines. There is an additional knowledgebase article spelling out some specific details. There were also two blog posts providing additional context that now redirect to the main Mojang page, but you can find archives of them here and here. The now-unavailable blog posts are substantially similar to the still-available Guidelines, and the Guidelines still contain links to those posts at the time of writing, so I am going to make the reasonable assumption - or at least I think it’s a reasonable assumption - that the content in the posts can be relied on. (ADDENDUM April 11, 2020: Mojang has released additional clarification about various things that you may not sell; please read it.)
This is a good time for me to add that I am a lawyer, but I am not your lawyer, and none of the following is legal advice or can/should be construed in any way as establishing an attorney-client relationship; if you need legal advice, you should consult competent counsel in your jurisdiction. If you can’t find a lawyer, call your local bar association and ask for a referral. Similarly, this is not attorney advertising, and I am not seeking clients related to the practice of law. In fact, my contract with my employer (at the time of publication) bars me from practicing law outside of that employment relationship. Oh, and I’m not affiliated with Mojang or Minecraft! These are my own, personal opinions and interpretations, and they could be dead wrong! If I end up hearing from Mojang, either with corrections or clarity, I’ll be sure to make a follow-up post.
The short, incomplete-but-material version of the Guidelines is that you can’t sell anything that affects gameplay or that provides one player an advantage over any other player. But they also go beyond that - you can’t even sell something that provides an opportunity to some players over others.
Practically, that means you can sell:
Cosmetics, like item skins (but not the items themselves)
Server-wide boosters
But not:
Items
Opportunities to obtain items (drop parties, boss keys)
Any kind of per-player currency
Access to commands
Access to gameplay systems
You get the general idea. If you have questions about specific applications, I can give my (non-expert, personal, non-authoritative, non-Mojang, unsanctioned) opinion about whether something would violate the EULA or not in the comments, or maybe a follow-up post. Quick yes/no applications of these concepts are the most likely additions to this post, whether by edit or by addendum. By the way, I have included GIFs every so often, because GIFs are great. I hope you enjoy them. And the post, of course, I hope you enjoy that too.
¹ “Large scale” here means large relative to my experience as compared to other platforms. I moderate and administrate Discord servers for a number of popular Minecraft plugins, and I have reviewed the monetization strategies of a very large number of Minecraft servers. One of the requirements for advertising on those Discord servers is that the posted server comply with the Mojang/Minecraft EULA; accordingly, I review the monetization strategies (if any) of each server owner that posts an ad. Many of them post these ads without even reading the rules. Shocker.
Approaches to monetization
I have a lot of opinions about ways that server owners monetize. I have opinions about whether these things violate the letter or the spirit of the EULA, I have opinions about whether they are morally permissible, and I also have opinions whether they are even a good idea from a design standpoint. I’ll start by outlining some popular approaches, then I’ll get into the particulars of each approach, and I’ll finish this section with some brief thoughts on each.
A shorter, more succinct version of this write-up would focus on the principles that apply broadly across categories, but one reason I endeavored to write this post was to provide it as a deeply comprehensive resource to server owners, so the analysis will be appropriately dense and focused. I’ll throw a tl;dr at the end, both for quick reference and to sum up what all I’ve covered.
The letter and the spirit, or: How I learned to stop worrying and love the EULA
Let’s talk about some popular monetization options and how they do or don’t violate the EULA. I’ll explain both the letter and the spirit of the Guidelines where applicable.
Server boosters
Server boosters multiply some positive aspect of the server for all players, constantly, for a defined period of time.
The letter of the Guidelines here is clear:
Therefore, if you comply with and follow the Brand and Asset Usage Guidelines and Naming Guidelines above YOU MAY:
…
sell positive effects or enhancements (other than “Capes”) IF everyone on the server is positively affected in exactly the same way.
A server boost is a “positive effect or enhancement.” Thus, a server owner may sell a boost that applies server-wide, but not a boost that only affects a specific player; “everyone on the server” must be “positively affected in the same way” - a purchaser of a server boost may not benefit more than the other players on the server. It must be a true server-wide boost.
Some servers have boosters that double the rate of rare drops when players kill monsters, or triple the currency that players earn from certain activities, or multiply the experience that players earn towards leveling up their account or their plugin-based skills. These are all EULA-compliant server boosts as long as they affect all players on the server equally while they are active.
The spirit of the Guidelines, in my opinion, is also clear. There is little debate about what “positive effects or enhancements” means. But “everyone on the server” and “positively affected in exactly the same way” are where some fudging begins to enter the conversation. Does “server” mean “world” if you separate your Bungee network into a different server instance per-world? (“A server is something a user connects to with their client;” no.) Does “exactly the same way” mean the same rate or the same general benefit? (The same rate, obviously.) Can you fudge the rates and give the purchaser of the booster a slightly higher rate? (See above; again, no, come on.)
To comply with the EULA, a booster should simply and unequivocally afford the same benefit for every player who is connected to the server, where “server” means “every interconnected instance of the game.”
Item cosmetics
Item cosmetics are cosmetic skins that can be applied to items without changing their underlying statistics, gameplay effects, skills, or otherwise. They change the item’s appearance or visual effects and nothing else. For example, some servers allow donors to have skins for their diamond swords that change the model of the sword to appear fancier and more elaborate, but do not affect the sword’s damage, range, attack speed, etc.
I think these are basically fine. I haven’t implemented them on my server yet, but mostly for technical reasons. (Prior to 1.14, the mechanics of doing so are a bit troublesome.) But they’re not terribly hard to create, and they can provide a lot of value to players.
Balancing item cosmetics typically revolves around balancing how the cosmetics are acquired and utilized, which is discussed below. The Commercial Usage Guidelines, importantly, do not state whether player may purchase item cosmetics for hard cash - USD, etc., - and then trade them to other players for server currency. I have a suspicion that Mojang would not be thrilled about this. My most cynical view - and I think probably the correct view - is that the tradable skins would essentially be “soft currency,” even if their value is subject to market conditions. The value of a dollar fluctuates, after all, and so does the value of any unit of soft currency in-game, whether it be “gold” in an economy plugin or gold blocks. Cosmetics, therefore, should probably not be “tradable.” Or, if the attached item is acquired by another player who has not paid for it, the skin should revert to its unpaid / original form. These are but a few ideas for how to implement cosmetics in a compliant fashion.
Loot boxes (a/k/a loot crates)
The bane of free-to-play players across the land! This post would be worthless if it did not mention loot boxes - it would be as valuable as the expected value of any standard loot crate in any given game that offers them.
Loot boxes are items - available for purchase, even if also offered for free - that award their opener a random prize as a roll on a table of options. Typically, loot boxes have some rare and/or powerful items available as extremely rare rolls, but they tend to award paltry prizes to the average unlucky opener.
Loot boxes in many games contain currencies and non-cosmetic bonuses that will give players some kind of advantage. Indeed, many Minecraft servers sell just such boxes. Doing so, however, violates the EULA. From the Knowledgebase article:
[Y]ou are not allowed to sell the following for hard currency:
…
Items or perks that affect gameplay or give the player an unfair advantage, even if they are also available through soft currency purchase, or just by playing the game for a certain time.
Selling an “item or perk … that affect[s] gameplay,” e.g. a loot box, is therefore violative. But loot boxes can be EULA-compliant! A loot box is compliant if it is “cosmetic,” per the Guidelines:
Therefore, if you comply with and follow the Brand and Asset Usage Guidelines and Naming Guidelines above YOU MAY:
…
sell cosmetic items, except for “Capes”, IF the item sold does not give a user an unfair gameplay advantage over anyone else on the server - i.e there can be no “pay to win”.
“Capes” are specific items that attach to the backs of players. Mojang asserts that they retain the exclusive right to award capes to players, and these are typically given out for events like Minecon. Read more here.
So if a loot box only awards “cosmetic” items, like those described above, it complies with the EULA. Popular server networks, including Hypixel, are notable for selling loot boxes with a large variety of cosmetics. (Disclaimer: I have some friends who work(ed) for Hypixel and Hytale. We previously worked for the same (non-Hypixel) employer.)
I don’t like non-compliant loot boxes. I think the ways that some servers have attempted to implement them are downright exploitative. They assert that they comply with the EULA because there is some twist to how those boxes are distributed, but the end result is still a violation.
Loot boxes that only award cosmetics are less onerous than their pay-to-win counterparts, to be sure. But, read onwards, because the road to hell is paved with loot box-related good intentions.
Thou mayest
Sure, there are all of these different ways to monetize your server. Some of them are permitted by the EULA. Some aren’t. But what are the moral considerations implicated by each approach?
Server boosters
It feels a bit unfair to start with server boosters. After all, they are probably the most morally-sound option of the bunch. Because boosters apply to everyone, regardless of who they are funded by, they don’t facially benefit specific players over others. But they’re not perfect.
While a server booster applies to the entire server, certain players will derive more benefit from that boost than others. Of course, boosters only apply for the period they are active, and players in different timezones may be disadvantaged if they can’t log on and play while a boost is live. For example, my player base is mostly located in the United States, but I have a few players who live in Australia and Europe. I built our automated weekend events so that every player on the server would be able to both contribute to the server-wide progress and benefit from the unlocked bonuses. The Australian players make some of the largest contributions!
Differences in how players benefit from a booster are also a function of what they’re able to accomplish. A veteran player who can generate 1,000 gold per hour makes 2,000 gold per hour with a booster active, while a newer player who can only generate 100 gold per hour will make 200 gold per hour while the booster is active. Because the booster doubles the actual returns of each player, income disparities will be exacerbated by it. These disparate earnings can create social divides between players, especially if a pattern develops where wealthy and powerful players are triggering boosts to further distance themselves from players who have less wealth both in-game and in real life.
Finally, consider whether the sale of a booster itself is morally neutral or not. Why does a player want to buy this booster? Do they need it? Is the game oppressive without it? Is the server balanced around an expectation that the booster will be active? Will the players progress through your content too quickly if boosters are constantly active? Can you still provide a satisfying experience? Are you selling something that is actually worth buying, or are you just putting shiny wrapping paper on a pure cash extraction? I’ll offer some of my thoughts about these questions, but you’ll have to answer them for yourself.
Item cosmetics
Item cosmetics present some sticky moral questions. Frequently, cosmetics are only available to the players who pay for them. So, when you’re selling a cosmetic, you’re selling a status symbol. Even if you provide a way for players to earn those cosmetics in-game, you’re still selling a shortcut. You’re letting someone pay money for the right to feel cooler and better than everyone else - or at least non-paying players. Selling item cosmetics is about much more than selling one cosmetic to one player.
Many server owners love to sell cosmetics because the items are walking advertisements for the server’s shop - in this way, one sale to a single player can lead to more sales to others, and the process becomes recursive. A cosmetic-less player who sees a cool sword, or a badass helmet, or whatever else - they might think to themselves, “I want that, and I’m willing to pay for it.” But not everyone can pay! So the player is then either out of luck - or they’re driven to grind for whatever server currency you sell the cosmetics for, often at the exclusion of efforts towards substantive (gameplay-based) progress. That currency, or the effort used to obtain it, would otherwise be diverted towards purposes that the player would perhaps prefer, but for the existence of the cosmetic.
That doesn’t mean selling cosmetics is bad. It just means that you’re introducing new layers of social complexity to your community - layers that are tied to spending cash, not effort. In a server without monetized cosmetics, the gear that players have represents their accomplishments. If you sell cosmetics, you decouple that relationship and introduce ambiguity. You also create new “goals” for players that may or may not be in line with your design philosophies. When it comes to item cosmetics, be sure to think about the impact they’ll have on your community, not just what you’re selling and how you’re selling it.
Loot boxes
The concept of a loot box has reached an almost mythical status in popular culture. Some countries have mandated disclosing the drop chances for each item in a loot box’s drop pool, some companies have gone forward making that information public even without a statutory requirement, and some locales have even banned the products entirely. A loot box is like a digital Gashapon, awarding whoever opens it with a prize that may delight or disappoint. Typically, anyone who seeks their most valuable treasures must obtain and open them en masse.
In many ways, loot boxes are nothing new under the sun. How many raids have MMO players run in pursuit of that one piece of equipment from that one boss to complete their build? How many hours have gamers spent grinding currency to afford that next shiny reward? How many times have they played a level over and over in pursuit of that last elusive achievement for perfect execution of a challenge?
But the loot box takes that grind and crystallizes it - in exchange for cash, not effort. A loot box is tangible; often times, and especially in Minecraft, it can be traded. This detaches the reward from the underlying effort. That rare boss-drop equipment, that expensive reward, that cosmetic otherwise locked behind an achievement - the rising trend is that where before you could only get these items from in-game efforts, you can now spend real money to buy loot boxes that reward them.
Two quick notes of import, though - first, to comply with the EULA, a loot box that can be purchased for real money may only award cosmetics. Second, the major issue with loot boxes isn’t just that the rewards they contain subvert player expectations, it’s that the rewards have low drop rates and thus require huge volumes of boxes to introduce them into the economy. (Whether or not they’re tradable, they’re still in the economy - see the above discussion about selling item cosmetics and loops.)
Selling loot boxes thus involves the additional layer of design complexity - and moral complexity - related to the drop rates of the items in the loot pool and the corresponding expected number of purchases necessary to roll any given item. Unlike selling cosmetics directly, the spending necessary to get a desired item can fluctuate wildly, and the mere act of purchasing and opening each marginal loot box may build unhealthy habits in your players. The Skinner Box is a well-documented concept, but many server owners are totally unaware of the psychological consequences of their design. Like Dan and James say in this Extra Creditz episode, “there is a lot wrong with the way our industry does loot boxes.” You should watch that episode.
How much does a dollar cost?
Having discussed whether various approaches are compliant and whether they are morally permissible, I want to finish off with some opinions about whether they are a good idea from an owner’s perspective. It’s true that some approaches can make you faster cash in the short run, and it’s hard not to look at those short-term earnings as the best strategy. Every choice comes with consequences, though, and if you really care about your server, you should make an informed decision that accounts for the long-term effects of the methods you consider.
Server boosters
Server boosters are my favorite way to monetize. A server-wide booster provides a reward for all players who choose to play, and a wealthy player can subsidize bonuses for others without directly disadvantaging any other player.
I think server boosters are a good idea, but under certain conditions.
The major problem with implementing server boosters is balancing both duration and effectiveness. How long of a booster should any player be able to purchase? Should there be a cooldown between purchases? Should there be a cooldown between activations per-player and server-wide? How big should the booster’s multiplier be? Is the server economy susceptible to disruption from large currency spikes? How does it affect long-term gameplay?
First, my opinion is that boosters should have long durations - like 24 hours - so that they don’t just benefit the purchaser. No matter when your players log on, they have an opportunity to make use of the benefit. It establishes that you’re timezone-neutral as to your player base, that anyone who wants to play is welcomed, and that their presence is appreciated. If I ever sell server-wide boosters, they’ll be in increments of at least 24 hours.
You should also consider that activating a booster is a prestigious activity and will afford the activating player some social capital. It might make some sense, then, to ensure that no single player is able to monopolize the opportunity to activate boosters. My rough approximation is that a healthy cooldown between activations is twice the length of the last activation - so if you’re using 24-hour boosters, you can personally activate one every three days.
But you might also consider that having a booster constantly active may have negative effects on the players’ appreciation of boosters generally. You might want to allow boosters to only be activated on weekends, or one week out of the month, to make them feel special and exciting when they’re active. In that case, I would modify the above cooldown to be twice the length of the last activation, but only counted during activation opportunities. For example, a player who activates a 72-hour boost over one weekend where the server only allows weekend boosts would not be able to activate another boost until the third weekend following the prior activation. Here’s a chart:
M T W R F S U
M T W R F S U
M T W R F S U
M T W R F S U
First weekend active, two weekends cooldown, third weekend active. If you find that there’s considerable competition among players to activate boosters, you might want to increase the cooldown period.
The reason to do all of this is to prevent boosters from affording permanent social prestige to any particular user. There should be a bonus, sure - the appreciation from other players towards the activating player is likely to help foster your server’s sense of community - but you don’t want anyone getting all despotic about the whole affair.
Second, I wrote above about how boosters can (and on some level, are designed to) exacerbate income inequality in an economy, and you shouldn’t take that too lightly. Income inequality is essentially a time-gated problem that you will have to eventually solve - because your economy is almost certainly subject to inflation, and because it is easier for rich players to make money than it is for poor players, you need gold sinks to balance off the gold taps and keep the entire economy from collapsing (or, less dramatically, from becoming hyper-inflated). (The structure and progress of economies might make a fun subject for a future post.)
You can account for wealth disparities by having your power gates be locked behind exponentially-larger requirements, and have your taxes scale accordingly. But you should note that these tools just draw out the late-game; they don’t necessarily solve the inequality problem itself. To manage your economy at scale, you need sustainable gold sinks.
Crucially, a booster will also likely result in a huge influx of currency to your server’s economy. Will the economy be able to handle that? Will players feel ambushed if an inflating gold supply means their items get bought up quickly? Can one player corner the market? These issues are fundamentally the same as the inequality issue, they just run on a shorter timetable. If your economy has any potential exploits, a server booster can make them quickly and readily apparent, especially if the boost is particularly large.
Finally, you should consider how server boosters affect your long-term gameplay. As I’ve described, they essentially cut down on the amount of time any given player needs to make progress towards any given goal. If your players are close to maxing out on the content that’s currently available, boosters mean you have to work faster to pump out new content. But if the content you’re releasing just starts the grind back over again, you’re going to exhaust your players.
Boosters can highlight issues with an existing development process; if you’re just racing your players to provide more content for them, or if they’re just chasing numbers, that will be worsened by a booster that lets players progress faster. And “if the most efficient way to progress becomes the least interesting thing to do… that’s a problem.”
But it’s also the case that, as a player, seeing bigger numbers feels good. It feels good to get loot more often, it feels good to get more of that loot, and it feels good to get paid server currency for the work the player is doing. Boosters pump all of that up! They’re a drug you don’t have to consume - they increase the amount of dopamine you’re releasing from gameplay because the good stuff is happening more often. The danger comes from keeping boosters active too often, in which case that elevated level feels normal. Or from setting the bonuses too high, in which case the “normal” levels will feel inadequate.
On the scale of things, and across the broad spectrum of monetization options, I like server boosters. I think it’s possible to balance well around them, I think players find them to be a useful and valuable option for monetizing the server, and the rewards they offer create a tangible satisfaction that doesn’t necessarily create a dependency. A well-designed booster offering is a fine monetization option indeed.
Item cosmetics
Does it really matter whether one player has a normal sword, and another player has a thunder sword that summons lightning every time they swing it, as long as they have the same underlying stats and mechanics? The obvious answer is “of course not, they do the same damage, who cares, let the money-havers enjoy their fancy toys and what-have-you.” The less obvious answer is more nuanced.
As I described above, when you introduce item cosmetics paid for with real money to your server, you’re introducing new layers of social complexity to your community. You are, intentionally or not, creating a social hierarchy and allowing your players to send social signals about status and power.
It may not matter to you whether some players have cosmetics and others don’t (check yourself on this, though, as it can become easy to bias yourself unintentionally towards players who contribute more to server funding), but it will almost certainly matter to your players. And not just to the players purchasing the items, but, critically, it will matter to the players who don’t purchase the items. If a meta develops around various cosmetics, players’ aspirations will shift in the general direction of obtaining those cosmetics. If they can’t get them, they’ll be disappointed.
What’s a responsible server owner to do? I have a few ideas.
First, you can make cosmetics obtainable both by spending money and by spending in-game currency (i.e. effort). As a reminder, the EULA forbids the following:
Items or perks that affect gameplay or give the player an unfair advantage, even if they are also available through soft currency purchase, or just by playing the game for a certain time.
Important to that clause is the phrase “affects gameplay.” The logical contrapositive of “you cannot sell items or perks that affect gameplay or give the player an unfair advantage” is “you can sell items or perks that do not affect gameplay and do not give the player an unfair advantage.” So! Until Mojang says otherwise, it sure looks to me like you can sell cosmetics - and only cosmetics - that can also be earned in-game.
The Halo: Master Chief Collection development blogs for October and November 2019 include some commentary and previews of how MCC’s seasonal rankings system will allow players to unlock cosmetic content. Halo: Reach was extremely well-known for its expansive customization options when it launched back on the Xbox 360, and it wouldn’t surprise me in the slightest if Microsoft sold a “pass” that allowed expedited access to various cosmetics. The same kind of system would be totally permissible on a Minecraft server! As a contrasting example, the Destiny 2 season pass system (archived) would not comply with the Minecraft EULA. I think it’s a balanced version of a season pass - the benefits are meager and the free version is still quite rewarding - but it does provide “items” and “perks that affect gameplay.”
You could also abandon the concept of a season pass entirely and just offer the cosmetics up for direct purchase, of course. The key to this first concept is that players can earn the same cosmetics both by playing the game and by spending money. Allowing players to earn the same cosmetics that can be bought with real money allows your players to penetrate the social hierarchy that cosmetics can create, even if they can’t afford to contribute to server costs.
Second, you can regularly give away cosmetics for free and disrupt the “market” yourself. As the owner of your community, you have the power to control every aspect of it. You can manipulate the economy as often or as little as you wish by altering the supply of various resources. Cosmetics are just another one of those resources!
Riot games implemented Hextech Crafting in League of Legends a few years ago. Players who perform well and behave honorably in their games are occasionally rewarded with bonus chests and keys, and opening a chest has a chance to reward the player with a skin shard. Those shards can then be crafted into skin permanents. When the player crafts a skin permanent, they own it forever - it’s as though they had purchased the skin directly from the store. Hextech Crafting is a great example for a Minecraft server owner because the benefits afforded by it are entirely cosmetic, but players can (and do) spend hundreds of dollars on those cosmetics. Player can purchase additional Hextech Chests at any time, and every few months Riot also sells “orb” packs that directly award skin shards in various bundles. Spending money on these is quite common, but even players who don’t spend a dime can still get an occasional free skin for a champion they like.
You could easily imagine replicating the Hextech Crafting system as it applies to Minecraft, allowing players to earn and craft components of cosmetic weapon and armor skins, etc. I’ll get into this a little bit more in the Loot Boxes section below, but the important part here is that Riot gives away cosmetic to players entirely for free, and prior to this update it was basically impossible to get a skin without having someone spend money on it. It increased the frequency of players using skins in-game, and it let players who can’t or won’t spend money still be part of the “skin meta” by investing their free earnings into skins for the champions they care about.
You don’t need a complicated crafting system to give away free cosmetics. Turns out, you can just… do it. Whenever you want. And you can give whatever you want to whoever you want. Incredible! My advice is that you occasionally give away cosmetics as part of server events, and to loyal players who would really appreciate a little extra flair. It’ll make a positive difference in your broader community, of course, but it will also just make them very happy. It’s good to make your players happy. That should be what you’re going for.
Finally, you can award cosmetics globally when various server-wide goals are achieved, regardless of who triggers the achievement. This could be a funding goal, or the completion of a dungeon, or the slaying of a boss, whatever. You get the idea.
I think this idea is particularly clever because it helps bring your players together in pursuit of a common interest. Players who may not be able to individually complete whatever task is at issue can still contribute to someone else’s quest for glory, and you can also design goals where every bit of effort counts. You could even create tiered awards that are cooler and more impressive depending on how much effort is spent by each player! These rewards are great because they can allow players to participate in the cosmetics meta even if they can’t afford to buy other fancy items.
You can also put a spin on the concept by improving the rewards gained by any individual player when they contribute to the success of another player. A direct incentive for cooperative behavior can help players build their relationships with each other and improve the health of your community! What’s not to like?
I think item cosmetics have the potential to be healthy and viable ways of monetizing your server. Like boosters, they require thoughtful implementation to ensure a stable economy and to foster healthy relationships between players. Be sure to think carefully about the impact of your cosmetics system and how it might play out once implemented, and don’t be afraid to fudge the market a little bit if you notice dangerous trends developing. It’s more important to keep the server strong and fun to play on than it is to squeeze out every last individual dollar - after all, those contributions will dry up if your player quit playing.
Loot boxes
Man, I love to hate loot boxes. So many games (and Minecraft servers) implement them as a way to hit-and-run their player pool. The other Extra Credits video about loot boxes and ethics is very good, and you should watch it.
Loot boxes, again, are only EULA-compliant when purchased for real money if they only award cosmetics and other EULA-compliant items. A loot box is just a wrapper for its contents, after all!
Good observation, Morty! Again, I’ll defer to the Extra Credits episode linked above for a detailed explanation about how loot boxes are and are not like gambling, but:
Loot boxes are like gambling in that they involve a consistent input and a randomized output.
Loot boxes are like gambling in that while there are expected costs and returns for any given set of inputs or outputs, actual costs and returns vary for each individual player.
Loot boxes are not like gambling in that you cannot “cash out” on the assets inside them; you cannot return them in exchange for their value because they do not have any inherent monetary value.
But couldn’t you say that monsters are loot boxes? You could say that creepers are loot boxes, that zombies are loot boxes, that the Ender Dragon and the Wither are loot boxes. They require different inputs and have different outputs, but they’re still loot boxes. Right? Well, not quite.
First, monsters and loot boxes are alike in the ways that they differ from gambling. Consistent input and randomized output, varying costs and returns per-player, no cashing out. But there are a number of stark differences!
Loot boxes are (frequently) obtained by paying money for them; obtaining loot from a monster requires the effort of slaying it.
The contents of a loot box (if EULA-compliant) do not affect gameplay; the loot dropped by monsters is fundamental to gameplay.
Loot boxes can be obtained en masse with no delay; loot from monsters is time-gated by the time it takes to slay each monster.
The ways that loot boxes differ from monsters are what make them particularly insidious. They allow a user to trigger those Skinner Box reactions in rapid-fire fashion, building up a dependency on the thrill of opening them. The rewards tend to be flashier and more culturally significant than the loot gained from monsters, and those rewards frequently have a more permanent and visible presence in the community. These attributes, when combined, can be extremely addictive - moreso, even, than just playing the game.
A good server owner should be concerned about balancing the habits of their players to avoid addiction in the first instance. You want to build a healthy experience, not a manipulative one, or you will ultimately cause burnout in even your most loyal players. Addiction when it comes to loot boxes is particularly toxic because it involves the spending of real money; this kind of addiction can have a far more material and significant impact on the real lives of your players. It is irresponsible, both morally and from a design perspective, to allow this kind of addiction to develop and sustain itself.
But, you know, a competent and deliberate server owner can implement loot boxes in a way that doesn’t ruin their server. And many have! Like I described above, I like Riot’s Hextech Crafting system; giving out free content to all players on a regular basis will help to level the cosmetic economy in small but meaningful ways. Another thing Riot does is place daily limits on certain gambling purchases during their Essence Emporium events.
The Essence Emporium allows players to redeem their gameplay-earned currency for some randomized cosmetic awards, and active players can amass truly mountainous quantities of the stuff. But when you go to redeem it for these randomized cosmetics, you can only pull the digital Skinner Box lever so many times per day. This is clever! It requires that players log in multiple times for the duration of the event to truly cash out on their currency reserves, and it also distributes the endorphin rushes from opening those random rolls across a longer period of time.
While there are certain limits on Hextech Crafting generally, a system I’d like to see Riot expand upon is limiting the bundle purchases during their pass-based events. The most expensive bundle sells for $50, and it’s easy to pick up two of them by dropping $100 without blinking an eye. This certainly helps fund the game and keep it free-to-play, but by pegging the best purchase value to such a large expenditure, it can cause some players to spend more than is healthy.
If you plan to implement loot boxes on your server, my advice is as follows:
The boxes should only contain cosmetic items (items that do not affect gameplay). This is required by the EULA.
There should be daily purchase limits on the boxes - something reasonably low. Your “whales” will keep coming back, don’t worry.
You should regularly give out free boxes. You can do this as a reward for logging in regularly, as a reward for certain gameplay accomplishments, whatever.
You should emphasize the contents of the box - including bad luck protection - rather than the act of opening it.
Loot boxes are best as fun, exciting flair that augments existing systems, rather than as a system with their own meta. When well-designed and carefully implemented, they can be a rewarding and unique way to distribute rewards for players on your server. But - and I cannot emphasize this enough - tread carefully.
Closing commentary
“So often, very little high-level thought is devoted to how the monetization experience feels to the player.” The linked video (from, you guessed it, Extra Credits) explains how monetization approaches that are not well-designed can undermine player experiences - not just on an individual server or for an individual game, but for the entire market of games writ large. “Build a good game and the money will come” - I can’t agree with this sentiment enough. If your monetization is complementary to to the gameplay experience itself, if it’s integrated into the core features of the game in a way that feels good and rewarding, even for the players who do not spend money, it will be successful and all of your players will be happy. The worst monetization designs are those that feel like an afterthought or an accessory; loyal contributors are left dissatisfied, potential smaller contributors may be deterred, and non-contributing players feel ignored.
Don’t fall into the Skinner Box pitfall. Don’t make your players spend money for short-term happiness that becomes long-term resent. In fact, you should avoid players feeling as though they need to spend money entirely. Monetization is most effective when it comes from the player’s own happiness, free will, and desire to support the server. That feeling is infectious. It bolsters your server’s success - both financially and on a community level - and it is the only truly sustainable approach to monetizing.
Video games are great because they let us feel like we’re part of something bigger - an epic story, a challenging quest, a bustling world. The concepts that are at the core of good game design don’t go out the window when it comes to designing monetization. If you are thoughtful and deliberate about designing your monetized content, your players will notice and appreciate it, and they will enjoy their experience all the more.
tl;dr
Monetizing your server involves a series of choices both before and during its implementation.
Do not discount the reality or the importance of each of these choices.
Do not prioritize short-term gains over long-term sustainability and growth.
Do not be afraid to alter your systems after they have been implemented if doing so will benefit your server and your players in the long run.
Focus your decision-making on principles of good game design.
Build a good server and the money will come.
Thanks for reading this extremely long piece. I may edit it after publishing, I will do my best to answer any questions, and it is entirely possible that I will write some addendums.
If you want to share this post, here’s that button.
I have some other topics I plan to write about in the future, so if you enjoyed this or found it useful, smash that subscribe button. Most content will be free - you can just ignore the part where it asks you to pay.
Finally, if you’d like to come play on my Minecraft server, we’d be glad to have you join us. I try to make interesting content, I’m proud of the community we’ve built, and we have a lot of fun. The IP is mc.summaxr.com, which also happens to let you view our live map when visited in your browser. Whether you join us or play elsewhere, I hope you found this information useful, and I wish you only the happiest of crafting.
Addendum - April 11, 2020
Mojang published some additional guidelines today about content, commands, etc. that may not be sold for hard currency (money). I added a link to it above, but I’ve linked it here as well because I want to be sure that nobody misses it. There is a useful bulleted list, but I also want to excerpt some specific content:
Hard currency is real money or anything that can be converted into real money, including, but not limited to, Bitcoins. Soft currency is available in-game only and has no real-world value. You may not sell soft currency for hard currency, neither separately or along with ranks as a single or recurring payout. This is regardless of what you can use the soft currency for.
Because the Commercial usage guidelines state that you may charge for access to your server only if you make a single charge per person that is the same for everyone, you may not sell “lives” or reduced ban time for hard currency. If you have a single entrance fee for your server, it needs to be the same for all players, including the ones connecting for the first time.
The Commercial usage guidelines also state that you may give everyone you charge access to all the Mods that you choose to have on your server, except only in respect of genuine admin tools/mods which should be reserved for administrators. Because of this, you may not sell “unbans”, “unmutes” or similar admin tools for your server for hard currency, regardless of who receives the money.
(Emphasis added.)
A new clarification here is that you may not sell soft currency for hard currency, even if you use the soft currency for cosmetic purchases. Please take note of that limitation.