Guest article: Game Commerce from virtual item sales to gold farming

The following Guest Article by Steven Davis is an extract from his recently published book, Protecting Games: A Security Handbook for Game Developers and Publishers. The extract is from Chapter 22: Game Commerce: Virtual Items, Real Money Transactions, Gold Farming, Escorting, and Power-Leveling

Money makes the (real and virtual) world go around. In some sense, money itself is the oldest, most widely used virtual item. Because money is so universally understood, virtual currencies are widely used as incentives in online games. Whatever one’s views are about “consumer culture,” we all seem to have a Pavlovian response to accumulating more things.

Game developers know this and draw on it to reward players. However, once you create a system where more is better, people respond creatively. Where there is a gap between those with more time than money and those with more money and time, someone will come along to close that gap. Welcome to game commerce. It is worth noting that game commerce is not applicable to all games after all, there are games of pure mental or physical accomplishment like Chess or Baseball. Although people do cheat in both, there is no way that most of us will ever be a Chess Grandmaster or in the Baseball Hall of Fame.

Anyone can be rich, however, especially in a game.

Game commerce encompasses legitimate transactions where players buy, sell, trade, and exchange items, skills, or characters; as well as unauthorized transactions. The most visible, and notorious, form of game commerce is gold farming, where players purchase currency or items for real cash and don’t earn them by playing the game. The other major categories of unauthorized game commerce include power-leveling, which involves hiring other people to play on one’s behalf, and escorting, in which players hire other people to play along with them as partners.

The problem for game developers is that, from a strict game play perspective, these activities are completely legal. In fact, many games that loathe game commerce are built expressly to support the very activities that make it possible. Game developers want players to be able to give items to each other and they strive to make it easy for players to group together with others and play the game socially.

The Dark Side: Four More Categories of Game Players

There are many reasons people play games. Dr. Richard Bartle proposed four basic
categories of game players (as modified to suit my purposes) [1]:

  • Achievers – Players who seek to maximize their score or items or status in the game.Explorers – Players who want to experience and understand the game world and its design.
  • Socializers – Players who use the game as a mechanism to form and expand their social circle.
  • Competitors – Players who want to compete with and excel over other players. Dr. Bartle seems to take a more negative view of this category than I do in that he includes abuse of other players as an implicit part of this category. To his four categories, I would add a set of mirror categories. The dark side of gaming:
  • Earners – Players who seek to earn the most wealth in the game for real-world reasons. These are the gold farmers and power-levelers.
  • Exploiters – Players who carefully study, explore, and analyze the game world and its mechanics to identify weaknesses that give them a substantial advantage, usually due to flaws in the game design and implementation.
  • Harassers – Players who use the social mechanisms of the game to make the experience as miserable for others as possible.
  • Dominators – Players who use the game’s mechanics to make other players miserable. These players are not really interested in doing better than other players, but in making other players know that they have been beaten.
  • Exploiters and Earners are often closely tied to game commerce. Exploiters help optimize the earning potential of the game. Harassers and Dominators cover two of the main categories of griefing.

As a game-design note, the existence of the four Bartle categories of players is probably one of the reasons that game commerce exists. Game developers are often Achievers and Explorers. They want players to work through the game and experience all of the developers’ carefully crafted content. The problem is that many players are not similarly motivated. Socializers want to be able to play with whom they want when they want. Some Achievers are more interested in status than achievement and they may not have as much time to play as the developers want them to devote to the game. Competitors are interested in competition, not resource gathering or exploration. Explorers may want to be able to go everywhere and do everything without “achieving” everything necessary to unlock all of the game’s doors. Game commerce is the shortcut for all of these players to achieve their goals in spite of the game designers’ wishes.

When game commerce is not explicitly permitted by game operators, it creates problems for the game operators because players will engage in game commerce activities, whether officially allowed by the game operator or not. In 2005, Nick Yee surveyed 1923 EverQuest players and found that 22 percent admitted to buying gold [2]. These gold buyers purchased an average of $135 per year in gold: fairly close to what they were paying in subscription fees to the game. Considering the fact that players are likely to under-represent the rate at which they do something that is frowned on, like gold buying, and under-report the amount that they spend on such items, it is very clear that gold buying is very widespread. Players pay for convenience.

Game commerce causes problems because it creates a mechanism where players can bypass parts of the game or the “effort” that the game’s developers, and a number of its players, feel that everyone should achieve (see the sidebar called “The Dark Side – Four More Categories of Game Players”). The other problem comes from when players engaging in game commerce activities interfere with day-today game play for ordinary players. This can range from monopolizing game resources in order to farm them most efficiently to broadcasting annoying advertising announcements for their services and clogging up the game’s communications channels (see Chapter 21).

The biggest problem with game commerce is that it is lucrative. Some estimates place total gold farming revenues at over $1 billion worldwide [3]. Unfortunately, this kind of money creates problems both within the industry and outside it. There have been a number of cases of employees using their access to online games to fraudulently create virtual items for sale. There is also the growing problem with online criminals targeting MMOs because of the ability to convert stolen account information into cash with little to no risk.

Amusement Park Economics

Many game-commerce problems are due to the simple, abstract economic systems that are found in many online games. Though these systems are called “economies,” they are really amusement parks. There is no supply and demand. Players “ring the bell and win a prize!” In some sense, problems arise because the games do not fully embrace their amusement park “nature”: assets can’t be stolen, but they can be bound to the player who picks them up; resources don’t have weight or size, but players have limited inventories; players can teleport or flying around quickly, but can’t take their characters to another game server or shard; the games are supposed to be “fun,” yet are designed around a treadmill or “grind” to force players to forever accumulate resources and repair “worn” items.

In some sense, these game’s economies are not really meant to be played for fun. Rather, they act as surrogates for the long lines and height restrictions that one finds in a traditional amusement park. After all, what is the requirement o accumulate resources but a way to delay players from entering “highvalue” instances or dungeons? And what are level restrictions, but ways to limit when a player can access certain adventures?

For many players, game commerce gives them a way to “cut to the front of the line” and “grow a couple of inches” so they can ride the roller coasters instead of being stuck on the kid’s rides. By this measure, gold farming, power-leveling, and the other forms of game commerce are symptoms of game design failures. After all, gold buyers and other game commerce consumers are giving money to someone else instead of the game operator in order to have the entertainment experience that they desire.

<h3Alternative Models

For a long time, the dominant business model for the online game industry has been “purchase and subscribe,” whereby a player would buy a shrink-wrapped game box and pay a monthly subscription to play. (It should be noted that before Internet Service Providers (ISPs) moved to flat monthly fees, many games and ISPs had a metered service where players paid by the hour or minute.) World of Warcraft, Lord of the Rings, and most other “major,” traditional MMOs follow this model with a standard retail price of around $50 for the game box and between $10 and $15 for a month’s subscription.

A number of companies have tried variants of this basic model. The most familiar alternative is to provide the game as a free download with a monthly subscription, as found in Eve Online and some of the older, larger Asian MMOs. Conversely, ArenaNet’s Guild Wars is purchased, like a standard game, but there is no fee to play online. No one else has really adopted this model for a persistent world game even though it was very successfully pioneered by Blizzard’s Battle.Net for the Starcraft, Warcraft, and Diablo games. A free online service combined with a purchased game is found routinely with first person shooters and real time strategy games. However, for these games, the online service is often little more than a lobby. Also, publishers have been quite willing to shut down the online service, almost on a whim.

Subscription games with a persistent world or economy are the primary victim of game commerce problems.

Over the past several years, the “free-to-play” (F2P) model has rapidly emerged. This business model is usually based on a game that requires no subscription to play, but collects revenue by the sale of individual virtual assets (the virtual currency is purchased in a number of ways, such as credit cards, debit cards, phone-based payments, wire transfers, or pre-paid cards). The F2P business model grew popular in Asia with the tremendous success of games like Nexon’s KartRider, MapleStory, and Audition. Recently, the F2P model has grown in Western markets because of the popularity of Asian games as well as the development of original game titles such as EA’s Battlefield Heroes.

Interestingly, some of the earliest adapters of the F2P were US-based text MUDs, such as Iron Realms’ Achaea, which launched with virtual currency purchases in 1997 and had its first virtual item auction in 1998 [4]. Iron Realms uses a dual currency system with one currency based on time in-game and the other based on direct purchases. The only type of trading that this system allows is between the game’s two currencies [5].

It should be noted that some games earn revenue from advertising or other marketing activities like surveys which are, in turn, linked to virtual item purchases.

There are seemingly endless variations on these models. Jagex’s RuneScape is free-to-play, but has a low-cost monthly subscription option that gives a number of advantages ($5.95 per month) – a strategy that is also used in Disney’s Club Penguin. Linden Lab’s Second Life’s primary revenue comes from selling virtual items and renting virtual real estate. Many F2P games still include an actual in-game economy just like the subscription games discussed here and are vulnerable to the same sorts of game commerce problems for this portion of their business.

Finally, there is the “broker” model – where an online game earns funds by brokering transactions between buyers and sellers of virtual goods and earns a transaction processing fee. Linden Lab earns some money from this model and IMVU does as well. In some sense, Apple’s App Store and iTunes use this same strategy. The challenge for the broker model is that the transaction processing margins need to be relatively large to handle any payment problems that occur, such as charge backs (see Chapters 27 and 28 on money and payment security issues) and bandwidth for large digital items. For Apple’s App Store, the company takes 30 percent of the transaction [6].

On Virtual Items

Who owns your virtual assets? The game developer or operator? The player? This question has vexed the online game and virtual world industries from the days of MUDs.

The general perspective of most developers is that the company owns all of the assets and that the player is simply renting access to them – much as they would a subscription to HBO. The assets and characters are non-transferable, have no intrinsic economic value, and can be altered or removed at the whim of the developer (as can the player).

There are excellent reasons for this. Game developers have been reluctant to invest in the effort needed to build highly robust, transactional, and reliable systems to store virtual asset information. Online games are in a constant state of flux and so the developers are also concerned that any alterations to the value of virtual assets (nerfing) might incur a liability for the company because players had already “invested” in those assets. Other questions include addressing what happens when the game comes to an end and issues related to gambling arise, as many online games include a chance element (see Chapter 31).

Others, most notably Raph Koster [7] and Erik Bethke, have argued for what many call an “Avatar’s Bill of Rights” [8]. Tony Walsh has argued for further “rights” related to data [9]. The term “avatar” is more than a bit misleading. They are really arguing more about the rights of players in online games and virtual worlds. They make many excellent suggestions from a best practices perspective and many of the issues that they raise are far beyond the scope of this book. A couple of issues that are relevant include the ownership of virtual assets and avatars and the rights of players in relation to banning and punishment (see Chapter 23).

The essential insight highlighted by the “Avatar Rights” movement is that players ascribe substantial value to their game characters and virtual assets. The willful denial of this fact has facilitated the growth of gold farming and criminals who target online games, in some sense. Because developers don’t consider the value that players put in their virtual “stuff,” customer service is often not responsive to player complaints about lost items. Also, the game systems are not built to easily log, track, remove, and restore these items in case of loss or theft. In some sense, this is ironic – the same game companies that argue vigorously that the virtual items have no value, at the same time are extraordinarily reluctant to restore players’ characters or virtual items after alleged theft. The argument is typically made that the players are abusing the system by allowing their items to be stolen (or, actually, selling them) and then making a complaint to the game operator.

If the items have no value, restore them.

However, this is not a matter of rights; it is a matter of good business. Bethke proposes a “Better EULA” [10], not some sort of formal and universal declaration of rights. The real key to this issue for online game businesses is to maximize their revenues and minimize their costs. To the extent expanded ownership of virtual items by players increases the popularity of a game or the revenues earned from each player is the extent to which online developers should extend rights and control of virtual items.

About the author

Steven Davis has over 20 years of IT and IT security expertise and has
focused on the security issues of the gaming industry for more than a
decade. He advises game companies, governments, and regulators around
the world. Mr. Davis has written numerous papers and speaks at
conferences on all aspects of game security. He is the author of the
book “Protecting Games” and the game security and industry blog, PlayNoEvil.
Mr. Davis has international patents on game security and IT security
techniques, most notably the anti-cheating protocols that underlie the SecurePlay anti-cheating library. He has designed several games including DiceHoldem and acts as a design consultant. He is currently the CEO of IT GlobalSecure which develops game security products and provides game security, IT security, and game design and evaluation services.


3 thoughts on “Guest article: Game Commerce from virtual item sales to gold farming

  1. The argument is typically made that the players are abusing the system by allowing their items to be stolen (or, actually, selling them) and then making a complaint to the game operator. If the items have no value, restore them.

    I think the problem here is that operators don’t want to get involved in policing transactions that happen on unsanctioned secondary markets. Restoring an item would mean that you would first have to establish that it has been stolen/scammed. You can’t just start restoring items every time someone asks you to do it, for obvious reasons.

    On the other hand, if the operator has a sanctioned secondary market, then scamming should not be possible since the exchange should be hardwired to the game database.


    "On the other hand, if the operator has a sanctioned secondary market, then scamming should not be possible since the exchange should be hardwired to the game database."


    Sanctioned secondary market does not necessarily mean that an operator facilitates the secondary market. In these cases the trading might not be hardwired to the system.

    If the secondary market was facilitated by an operator, it would not directly prevent hackers from "scamming/stealing" items and therefore the same situation remains, where the operator would have to establish that an item has been stolen etc… Facilitated secondary market does, however, diminish the market for gold farmers and such.


    "I think the problem here is that operators don’t want to get involved
    in policing transactions that happen on unsanctioned secondary markets.
    Restoring an item would mean that you would first have to establish
    that it has been stolen/scammed. You can’t just start restoring items
    every time someone asks you to do it, for obvious reasons."


    The reluctance to restore items is also due to maintaining balance of in-game economy and saving in customer service costs. It has probably been wise to make a preemptive decision how to deal with these cases. While restoring single items might seems like a little and harmless way to deal with these problem, the process of doing it, including customer contact, problem solving and tracking, proving the loss, etc, is very costly.

    While the effects of single item restorations to in-game economy might be minimal, accumulated effects might have an impact. The decision not to restore stolen items prevents from items popping into existence outside designed mechanics, unless the stolen items are tracked and deleted.

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