The Protest Fallacy: Unsubscriptions and design decisions

He hasn’t worked on World of Warcraft in years, but I still miss Greg Street.

The reason I miss him is that he had a habit of succinct and cogent explanations of why Blizzard did things in WoW, and although recently we’re starting to see more of that again, we’ve never had that level of dialogue since his departure. Even now that he’s well and truly ensconced over at Riot Games as League of Legends design director, he still has a lot of insight to share on game design in general, and massive games like World of Warcraft in particular.

This recent post on his tumblr is chock full of such insights. One of them is about game boycotting, the often advocated process of voting with your dollars when it comes to a change you don’t like.

Dropping a game because of a specific design change (despite what you might read on forums / Reddit) is actually pretty rare. I know it happens, but if you’re stack ranking the reasons why people quit, those specific responses end up being so far down the list that it is hard for a development team to take actionable feedback. It’s really rare you see “Wow, that change we made cost us 10,000 players. Let’s revert it!” So overall, I would not advocate boycotting a game as a way to make a statement, especially if deep down you still love the game. You’re just not likely to drive change as a result.

This is fascinating to me for a number of reasons. For a start, it’s contrary to accepted wisdom that “the only thing devs/companies respect is money” because it is apparently so rare that it can’t actually have that much of an impact in the first place. If you unsub over a design choice, the only thing it means is you can’t play the game anymore. There is essentially not much else that happens. Devs care about these things, but there’s not a lot they can actually deduce from such decisions, entirely because they are so unlikely to happen.

Why we quit

According to Mr. Street, the most likely reason for players to unsub from a game of such size and duration as World of Warcraft (a game that at times has millions of subscribers and has lasted for thirteen years) is either their group of friends no longer plays, or they themselves no longer have time to play due to various life issues. Again, both of these are difficult for devs to really address in their design and maintenance of the games they steward. The easiest fix to any drop in subscribers is to innovate and shake up the design, but even that needs to be done carefully for fear of making the game too difficult for returning players to pick up when they do return. I’d also assume there’s at least some risk of alienating current players with radical departures.

One of the things we’ve noted over the years as players is that radical design changes, when they do come, almost always come with a new expansion rather than during a current one. This is likely a driving motivating factor behind that tendency, as players are more open to and accepting of large scale design changes in the period before and during a new expansion launch, when they’ll have the leveling process to use as a learning period. Dropping a radical redesign in between raids would be jarring for current players, after all.

What actually works

But does this mean that protest unsubscriptions are pointless? No, I wouldn’t say that. If you’re not enjoying a game, why should you pay to play it? If a change is so drastic that you simply can’t continue to enjoy the game, then paying for it is foolish. People unsub for a variety of reasons — a lack of content, greener-seeming pastures over in some other game, the aforementioned lack of friends and/or lack of time to play, a feeling of dissatisfaction with design choices certainly ranks as a valid reason. And it isn’t that a sufficient number of protest unsubs wouldn’t work, so much as in general the people advocating and enacting such are a small, vocal minority and their actions are not generally as significant as they may think. If a hundred vocal forum posters unsub over a change (say, flight in Warlords in Draenor) the people who frequent that forum may believe it a huge decision. But compared to the numbers of people who unsub for other reasons, it’s a drop in a bucket. MMOs churn, it’s a long established fact of the genre — players come, go, and come back all the time.

So it’s less that protest unsubs don’t work and more that they generally don’t happen, and when they do, they’re a very small proportion of those that unsub. Do developers care why people unsubscribe? Of course they do, that’s why they collect the data on it in the first place. This means that when you unsub, if you want your decision to be noticed by the developers, you need to make sure you take the time to inform them.

Talking vs. yelling

If protest unsubs don’t work, what does? Well, generally, talking to them does a trick. In the case of World of Warcraft, they really do pay attention to the forums, whether it’s the official ones or other forums where the game is discussed. Reasonable and thoughtful discussion is far more likely to get noticed that frothing rage, of course. Nobody likes to be yelled at. There’s a tendency to feel ignored because dozens of posts are made on the forums and nothing seems to be changing, but the fact of the matter is, with a game this size and this complex you can’t always just flip a switch and fix a problem. Sometimes designing the solution takes time, and sometimes the cost in players who dislike a change is offset by players who like it. That doesn’t mean you’re being ignored.

I’d love to see someone on the current WoW development team take a whack at discussing what kind of player feedback they most want and how they approach protest unsubs. WoW has had some significant periods of player dissatisfaction (again, Flight in Draenor comes to mind) and I’d love to hear from them about what goes into their decision making process when it comes to design changes.