Building frustration

Frustration is what drives players to actually spend money in a "free" game. The mechanism is simple: 1) get the players to enjoy the game, 2) give them a taste for game progression through short and medium-term rewards, 3) make new progression rewards increasingly numerous AND long to get. When they cannot wait any longer, they'll start buying. There are variations around this principle, but you get the idea.

That sounds awful -- and it is hard to imagine that any gamer, in particular demanding hardcore gamers -- would break down and buy something.

But they do.

A successful freemium game is designed in such a way that the player will never resent that situation. How is that possible?

First, the game must offer genuine gameplay depth and be generous. If players feel that the game is asking for real money very early on or is offering its full features only to players that have paid, they'll quit. But that's not all; you want the player to get lots of rewards first and make more advanced ones increasingly long to obtain.

I had to face that challenge while working on the design of Kartoon, a freemium racing game currently under development at Kadank. To get players excited, I built a leveling tree that would unlock new features (new vehicles, new parts, etc.) in rapid succession first, then would make it increasingly difficult to get new ones.

If players were to unlock new features too fast, we would run out of new ones very soon, and the player would have nothing new to look forward to; but if we allowed too much time between feature unlocks, we would take the risk of losing players. Thus controlling the leveling pace was key to the monetization strategy. How did I proceed?

First, we decided to use an energy system to limit the number of races a player could complete every day. But it was not enough to estimate how many races a player would run, and how much XP he would get for each race. Our game system was designed to reward the player's performance, not his assiduity in playing the game.

That is a key difference compared to freemium games targeted at casual gamers; those games reward players for simply playing, not for their performance. Since we were targeting gamers, we had to reward skill. In order to estimate how much XP a player would get for every race, I defined three gamer profiles -- light, medium and heavy -- that differed in the number of races they would run every day, and the amount of XP their skill level would yield. Then, I tuned the various parameters in order to make sure that the players belonging to those three categories would level up at the pace we had decided.

The challenge was that game values are the same for all gamers. In order to solve that equation, I experimented with different values until I reached the following curves. Each point denotes when a player for a given profile would reach a given level. For instance, a heavy player would reach Level 5 roughly during his fourth day of game. The three curves show that all profiles level up fast at first, then at a slower rate, but without too much time between two increments. The graph was a key tool to visualize the impact of a parameter modification on the leveling curve of all gamer profiles. This way, we could initiate the frustration loop.

So, which the techniques are used to generate this frustration, and how effective are they?

Limited action. The player is limited in the number of actions he can do. To do more, he needs to wait for some action resource to replenish. This mechanism is heavily used by Zynga but is avoided by nearly all hardcore-focused freemium titles.

Mandatory time. Key actions, like building new units or harvesting resources, take time, sometimes several days. This technique is heavily used by Kabam for its hardcore strategy titles and is used in Ubisoft's The Settlers Online.

Leveling. To unlock new features, the player has to level up by accumulating XP and coins, which take time to earn. World of Tanks is a good example of this strategy but we can find it also in EA's Battlefield Play4Free.

Which technique is best adapted to hardcore gamers? I believe it is the last one. Hardcore gamers tend to play a lot; therefore the first mechanism -- limited action -- would frustrate them right away. Mandatory time is better. It requires the player to launch his game quite often in order to harvest his resources or to initiate a new construction cycle. Frequent, but short, game sessions foster addiction.

This is seen as a good thing, from the point of view of the game publisher, but I still find that mechanism too frustrating because it kicks in too early in the game before you have the time to get "hooked" on it. In my opinion, leveling works better and is easier to tune than the two previous mechanisms.

Could a freemium game work without frustration-generating mechanisms? Probably not. Battlefield Heroes is an interesting case study. Ben Cousins, former general manager of EA's free-to-play Easy studio in Stockholm, reported that initial monetization was quite poor: "metrics showed that free currency earned through play allowed players to maintain multiple valid characters for totally free." Things started to change when the team made it more difficult to access everything, thus generating some form of frustration. EA learned the lesson, and Battlefield Play4Free, which came later, relies heavily on leveling to unlock items and to upgrade the player's soldiers.