25 July, 2006
“What the customer wants is better products for free,” – Dilbert
No, that isn’t an argument using an advertising-based business model in online games. In the Dilbert comic strip, the PHB claims they are going to start listening to the customer, and asks what do customers really want. Dilbert answers with the line quoted above.
In other words, the customer isn’t always right. Shh! Don’t tell them that; it’ll be our little secret.
But, this leads up to a question that most good developers think about: how can you get good feedback from the playerbase?
The main problem is that you can’t do either extreme: You can’t ignore every customer because some people have good feedback. On the other hand, you can’t listen to everyone because there is simply too much information. Even if you have layers of community managers reading and filtering content, there’s no guarantee that they’ll be able to pick out the truly important bits.
Ryan Shwayder posted about influential people on his blog. These people are also known as social network hubs, people who are in contact and that influence many other people. As Ryan points out, if you can keep these people happy their positive latitudes can influence others. However, if they are unhappy they will not hesitate to tell other people and let them know their grievances. It’s a good idea to keep them happy.
But, the problem is: how do you tell who are your hubs? A great way would be to find people who have done research and hire them on. Social network research is relatively new to online games, but the social network study has been done in the “real world” as well. Applying this information to online games is even easier, since we can track anything that happens in our games.
The problem is that you can easily misidentify the social hubs in your game. Do the people posting on your forum count? What about guild leaders? What about crafters? Unfortunately, there’s no single attribute you can identify to say, “this person influences other people.” Even hard-core socializers might just enjoy chatting and not have opinions on the game, so they may not influence other people in ways that the developers care about.
The other problem is that you really have to process the feedback to be useful. One important thing to consider is that people who like something are not necessarily going to be vocal about liking it. It’s easy to see a lot of complaints and assume something is wrong. If 20% of your playerbase is on the forums complaining about something, you should change it, right? Well, not necessarily. What if 50% of your playerbase actually like the current system and would hate it if changed? My rule of thumb is that for every change, about 25% will love it, 25% will hate it, and 50% won’t care one way or another. You’ll likely only get a fraction of the people that hate a change to actually complain about it, and an even smaller faction of people that like the change to publicly give support. The ratio of people who will become vocal will skew towards the negative side even more if people can apply social pressure to “dissenters”. In Meridian 59, people who supported our decisions often made their support for our changes known directly to us. They didn’t want to be singled out and PKed by the people opposing the decision!
Unfortunately, this leads to more conservatism in game design. If people are complaining about the game, that means they are still playing it. The people complaining aren’t the only people you have to worry about: the harder cases are people that just quit when a new change comes along that they do not like. The ultimate measurement of success in a game is the number of people playing. A change could make people happier… or it could piss off enough people to cause a drop in revenue.
The other big problem is the echo chamber situation. People who want to complain about something will go someplace public to complain. Other people with similar complaints will go to the same place and back up the complaint. Soon it seems that “everyone” is complaining about this problem, reinforcing each other’s complaints, and the developers aren’t doing anything about it. Even if “everyone” is only a dozen people making multiple posts on a single thread on the forum. Online forums, particular game forums, can be a place where opinions are reinforced by others. Keep in mind that only a small fraction of players in a game actually go visit the forums in most games, so relying on the forums for all your information can be problematic.
In addition, sometimes the player doesn’t even know what they want. A good game designer is creative and will come up with creative ideas that surprise the players. Before The Sims came out, how many people wanted a game where you could control how often a virtual character could visit the bathroom? I certainly didn’t think I wanted to micromanage someone peeing, but Will Wright showed that this can be part of a rather fun game. The sales figures back him up: The Sims was incredibly popular, even though no one was specifically asking fo this type of game. My friend and business partner Rob “Q” Ellis II often says that our job as game developers is to give the players what they didn’t even know they wanted. I think that’s wise advice.
You also have the problem that some people just like to complain. Everyone complains about X so you pull a week of all-nighters to fix X. Great! Now Y is the one thing that is wrong with the game and it would be soooo great if only Y were fixed. Another week of all-nighters and you fix it. Wonderful! Of course, there’s this problem Z that has been plaguing the game forever…. Online games are dynamic, and there’s no such thing as a state of perfection (regardless of nostalgia tells some players!). You will always have to change the game in one way or another, and therefore will always have people complaining about something that isn’t exactly as they like it. I think you could call this “structural complaining”, after the concept of structural unemployment. Some amount of the complaining will simply be present because there’s no way to make all the people happy all the time.
And, sometimes the complaints simply cannot be addressed. With the area I live in hitting unseasonable 100+ F temperatures, I’ve been seeking the comfort of restaurant air conditioning. Of course, when my food comes it’s hot, which ruins the cool air for a bit. It sure would be nice if they could cook my food without using heat! While it is possible to fix food without heat, this is perhaps a bit unrealistic to expect from the local In-N-Out. Just because I’m a paying customer doesn’t mean I can demand they find a way to cook my french fries without frying!
Keep in mind that with a PvP-focused game these complaints will only get worse. People want any advantage possible, and sometimes convincing a developer is about the best advantage you can have. Small changes will result in large complaints when people think that they are losing power over others. It’s very important to consider feedback very carefully before doing something knee-jerk.
However, don’t think the entire problem rests with the players. There are quite a few issues that cause developers to gum up the system as well. It’s well known that developers are hesitant to throw out systems even if they aren’t working quite right. Of course, sometimes the system might working just fine, but the players aren’t happy with the results because it restricts some undesirable behavior they would like to engage in. In addition, if a major change doesn’t work out, or in the worst case actually results in players leaving, guess who gets fired? Hint: not the players agitating for the change. So, again, you see some reasons for conservative behavior come in.
Developers also tend to overreact to feedback and make the problem worse. Is X too weak? Well, we’ll double the power and now X is certainly powerful! Well, yes, but way too powerful. So, you have to nerf X again; even if it’s more powerful now than it was before, people will only pay attention to the nerf. The most common way to overreact to feedback is to make 2 changes. Is spell X “useless”? Lower the casting cost or increase the effect, but do not do both! If you do both you will almost certainly find yourself on the receiving end that X is too powerful now and must be nerfed! (Usually by people who don’t use spell X….)
You might also have to accept the fact that not every customer is desirable. If I go into In-N-Out and loudly demand that they cook my hamburger without heat, they will throw me out of the store. They will determine that my custom is not worth the effort they have to go through in order to satisfy me. Likewise with the players of our games. The people who complain about having to spend time preparing before they can PvP may not want to play an online RPG. They might find the online play of FPS or RTS type games to be more satisfying; you have to spend less time preparing for a conflict and more time actually playing the fight out. Trying to change an online game radically into a different genre generally isn’t a good solution. Trying to pander to every single player can lead to more headaches than income. The internet is all about getting people what they really want, exploiting the niche interests. Perhaps it’s time we consider that for our own games?
What do you think? Try to keep the comments polite, or I will edit them. I’ve only had to do that a few time in the past, so don’t make me start now.