Piracy is an ever-present battle for game developers because we make products that can be duplicated at almost zero cost. Our livelihoods therefore depend on the honesty and generosity of our fans.

A lot of development time (and money) goes into anti-piracy measures for video games, but is it even worth the effort? Is piracy, as many pirates claim, a victimless crime? In this article I briefly outline the numbers and logic behind the decision to fight piracy.

One of the most flame-war-inducing questions related to piracy is why people do it. Undoubtedly much piracy is done by people who can't afford what they are stealing or don't have access to legitimate forms of the game (e.g. due to country-specific distribution issues). Certainly some theft is done by people wanting to "stick it to the man" or, amusingly, as a form of revenge against draconian DRM policies.

But none of that really matters to the people who are trying to eke out a living selling games (i.e. us). What matters to us is if we can convert thieves into customers or, when we can't, if nonconvertible thieves are actually detrimental to our livelihoods.

Lost Income

First things first: does there exist a sizable population of "convertible thieves"? That is, are there many players who are able to buy the game but instead choose not to because it is possible to steal it? If so, then those players represent true lost income for the game developer. Making theft difficult might then take those would-be pirates down the alternate path of becoming customers.

At the moment, our successful titles are on mobile and are free-to-play with in-app purchases. Because in-app purchases are implemented by the games and app stores together, instead of by the app stores alone (as with pay-up-front) the game can attempt to detect and block fraudulent purchases. By collecting the related stats, we end up with data that directly answers the question of how much lost income we would have if piracy was allowed.

The in-app purchase stats from Quadropus, Flop Rocket, Towelfight 2, and Roid Rage are below. The data cover all 30,000 members of our BscotchID community and over 2 million anonymous installations. The two app stores (iOS and Google Play), as well as each individual game, have widely varying numbers when it comes to purchasing rates (whether fake or real). And so I won't name names, because there are many potential reasons for the differences and it would be disingenuous to imply that a higher number for one game or store are due to properties of that game or that store alone. We don't have all the variables, we can't know, so why start a flame war?

Anyway, sorry, where was I? OH YES, the data. The ranges represent the range of values across all our games and both distributors. Bscotch, as opposed to anonymous, refers to being logged into BscotchID at the time of "purchase".

  • 50-70% of people making anonymous purchases attempt to do so fraudulently.
  • 22-34% of people making Bscotch purchases attempt to do so fraudulently.
  • Most players (66% of Bscotch players, >99% of anonymous players) do not make any purchase attempts at all, real or fraudulent.
  • 1-5% of players making real purchases are converted thieves. They first made a fraudulent purchase attempt, but after being blocked from doing so, made a real purchase.

What do these numbers tell us? First, as is common knowledge in game dev, most in-app purchase income is derived from a tiny fraction of players. However, it is important to note that a potentially very large fraction of those non-paying players don't even like your game in the first place. The relevant stat there would be how much time paying versus non-paying players spend in your game, which is a stat I'll cover in a later post.

Based on the fact that players who log in are dramatically less likely to attempt to steal from us (by a factor of two, roughly), one might be inclined to believe that logging in is itself a deterrent against piracy. This may be the case, but it could just as well be the other way around (players who are going to pirate your game are less likely to log in). Correlation is not causation, after all.

Getting back to the question at hand, players who attempt to steal from us (but can't) make up a substantial fraction of our players (more than 50%, taking into account all of the data) but only 1-5% of our income is derived from blocked pirates who convert into paying customers. When theft is taken off the table, pirates have two options: paid access or no access, and they overwhelming choose no access.

In other words: pirates gonna pirate.

Consequences to Developers

Only 1-5% of our income comes from converted thieves. That fraction could be written off as the cost of doing business, since gaining those converts requires developing and maintaining some form of fraud prevention, which can be a costly exercise.

But lost income is not the only metric that has to be considered. Studios with per-user costs over time, common with web-enabled games, can spend a lot of server resources on players who are free-loading on the generosity of others. Most purchases are fraudulent and, consequently, thieves make up the bulk of daily server expenses if they are allowed to use your services. It is not at all hard to envision the scenario where web maintenance costs rise above game revenue for highly pirated games.

Lost income and per-user costs are values that can be estimated with decent accuracy, if you've collected the data. But there are other aspects of piracy that are less tangible. Many discussion on the topic focus on fairness. It is upsetting, as a game developer, to have people steal your work while you barely scrape by financially (or don't scrape at all). It is upsetting, as a paying player, to see that everyone around you didn't have to pay for the same game. It is upsetting, as a pirate, to have to pay for things you want (especially when purchasing isn't an option, or the impact of even a small purchase on your income makes doing so financially unwise).

We don't want to spend our time in an arms race against thieves. The one thing we do have control over is web-enabled content. And so for us the solution is the tried-and-true one of developing online content that makes our games even better (and thus more valuable) but that is exclusively available to players with verified game purchases. This way we provide a powerful incentive for convertible pirates to become customers (since they'll end up with a better gaming experience) while preventing our server costs from being lost to thieves. We'll have done what we can, in a way that is fair to all parties involved.

The piracy numbers can be overwhelming and discouraging, but remember that most of it doesn't affect your livelihood. And when it does, there are things you can do to limit its effects. So, yeah, pirates gonna pirate, and devs gotta dev.