Wednesday, October 19, 2011

How to Draw Players into Communities

I've touched that subject before, but it certainly deserves its own post. First let us all agree on this: Communities are good for MMOs (actually, any product). Zynga would agree, CCP would agree, EA, Activision/Blizzard, .. anybody who says 'no'? Good...

But how do we design a game that makes players create communities? First, we need to give them some tools. For example guilds. That's already enough to make a small percentage of players create a community. And that's why virtually every MMO I ever played had these tools. But it's not enough.

The majority of players doesn't really want to join a community. And I can really report from own experience. I knew that I would have more (any?) fun playing Eve Online if I joined a community. And after a long time I did join one (arbitrary) corporation eventually. But then I didn't really 'like' the people there. It's not that they were unkind. I just didn't really like playing with them. They were talking about things I didn't consider interesting. And they did so in a way that I wasn't accustomed to. So, because I didn't want to leave them so soon after I had joined, I just stopped logging in. From a game designer's point of view this community magnificently backfired!

I don't really enjoy joining communities. And most people do not. We absolutely enjoy our existing communities, but we don't enjoy any effort to become part of a new one. And one reason is that becoming part of a community can absolutely not work out. Like in this example.
Of course, there are counter-examples. Some people absolutely love to create new friendships. I have a few real life friends who do it constantly. But they are a minority. Most people don't like creating or joining communities. They just enjoy being part of one.
Consequently, just giving the players the tools to create and administrate a community isn't enough.

Also, tools to find the right community for the player, are not enough. Eve has them, and now WoW has them. But, really, do you enjoy using the Guild Finder? Sure, you might like the result, but using it isn't that much fun. And one reason is that, while it's certainly of interest what activities your future guild engages in and at which weekdays, what is at least as important is whether you actually like the people over there. If you end up not liking the group of people you joined, the whole community aspect can absolutely backfire.

The second biggest problem with these tools is that they only help those players find a guild who actively look for one. The new players who don't actively look for a guild - some of them are afraid to interact with any stranger - don't benefit from from these tools at all. And, make no mistake, these players are often a silent majority! Moreover they, being new, haven't played the game for long yet, which means that they potentially have many recurring subscription periods in front of them.

This reminds me that I put Gevlon back on the blogroll. He did have some insights into WoW raiding lately which are worth reading. Even though I don't agree with many of his conclusions, he does analyze many problems correctly.

Next, you can give existing communities incentives to recruit new players. Eve, as well as WoW nowadays, do this. And it's not wrong. But it's still not really good either. This is how it works with me: "Pimpmix has just whispered me to join his guild. ... .. .... no".

Also, these systems can have side effects. Just look at guild achievements in WoW that made many smaller guilds dissolve, because players joined bigger, more anonymous guilds.

All these tools and ideas aren't wrong, but the best way to get players into communities is this:

You offer content that can be enjoyed alone. But you also make sure that it is slightly more efficient to group up. Make your players meet each other while doing the same thing. And make sure that they meet each other repeatedly. This way they will start to empathize with each other. They start thinking: “Hey, this guy has been here yesterday, and the day before. He is similar to me.” Yeah, you manipulate people. It's not always bad. In this case it is for their own good.

After being engaged in the same activity for some days / weeks, the players will eventually start to talk. It will probably require some kind of event for this to happen. For example, one player notices how the other one is in trouble after an unlucky pull and helps him out. The other one says “thanks” and so a conversation is started. A few moments later they group up to farm the mobs together, because it is slightly more efficient.

But even annoying things can help. Just yesterday I was on a LFD run with my rogue. I selected “need” for a locked box, like I always do. In my experience, non-rogues sell them unopened anyway, and nobody ever said anything. But this time the healer with a female avatar (that always plays a role, yeah, silly) complained. I explained myself and apologized (I am a nice guy as soon as you start to know me. It's just before we talked that I don't care). We ended up doing three more instances together whispering about the ridiculous tanks we got. Then we broke up, because, well, it's not like I will ever see you again, anyway. bb.
So, even if one char 'steals' some resource node from the other or even causes the other one's death, this isn't necessarily bad for the game.

I got to know the first guild I ever joined in WoW, and that I ended up main-tanking for two years later in TBC, by repeatedly meeting the same hunter and his shaman friend in instances while leveling. Eventually he challenged me for a duel. As an inexperienced mage vs. a hunter, I lost many times. I tried to keep distance, stupid me. Yeah, I even remember! I also remember about a year later me challenging him and beating him repeatedly with my new rang 12 armor set and by starting the duel inside his safe zone. haha! ;)
Anyway, this got us talking and ultimately me joining his new guild. God knows how long I had played WoW had I not joined his guild back then!

If you remember my first example about how I ended up not logging into Eve any more, you see why it is superiour game design if players talk to each other about something else than joining the guild, before they start to talk about joining the guild.

Summarizing, you want people to meet each other repeatedly while doing content on their own. You want there to be occasions that make it socially awkward to not start talking. You want there to be a slight incentive to team up now - and in the future.

And that's really all the magic behind it.


  1. "Summarizing, you want people to meet each other repeatedly while doing content on their own. You want there to be occasions that make it socially awkward to not start talking. You want there to be a slight incentive to team up now - and in the future.

    And that's really all the magic behind it."

    How can you achieve this in a moba game?What you describe here is only viable in a game that enphasize in simulation and in virtual world, but those two things are know "blacklisted" as grind, waste of time, korean style games

  2. @Giannis: Same as in WoW? For example, LoL allows you to add people you played with recently into your friends list, and then play with them instead of randoms.

  3. Nils, have you read Koster's essays and postmortems on online worlds?

    After reading these and looking at the current market, I came to the conclusion that we (the players) threw the baby out with the bathwater. Players wanted accessibility, convenience, and as little downtime as possible, but these are the antithesis of the very world the players are in!

  4. Some of the things you propose are impossible in the current game market. For the simple reason that one competitor may decide not to do them and end with a game which is more rewarding in the short term. Then all you get is people leaving.

    Having the same people doing the same content is extremely hard in a level-based system. Either leveling goes too fast, and a small difference in play time separates them, or leveling goes too slow, and then a difference in play time slowly creates a chasm which takes very long to fill.

    BTW reading your text, I would say that the LFD daily heroic is a good approximation:
    - it requires a group, but not necessarily a lot of chatting (some called it "solo playing with 4 bots", I think).
    - there are no level/item level barriers (almost).
    - if you find good players you can add them to your friends list and see if tomorrow, same hour, they are around for another dungeon.
    - it has downtime (regen, resurrect, regroup before a pack/boss).
    - it's not a repetitive activity like farming tons of mobs (even if this last one is better for downtime, grind is currently bad in MMOs, thank Elune).

    Honestly, I don't think it's so simple. In particular it's not so simple if the players, from the start, DON'T want to socialize.

    BTW I went to see Gevlon's blog, he seems to be as wrong as usual. Or, to say it differently, it's not analyzing anything, just showing that social is bad, facts be damned.

  5. I like some of your arguments, but some of your assumptions are just bad. For example, all players don't hate finding community. I would actually argue otherwise. You say that you don't enjoy joining communities and then you extrapolate that to "the majority" of players. This isn't necessary to make the points you're arguing; it can't even be proven.

    In fact, the prevalence of guilds in MMORPGs suggests something completely contrary to this.

  6. For example, all players don't hate finding community.

    I believe Nils has a defensible argument in this regard. "Guild hopping" has a negative connotation, for example, even if the reason happens to be things simply not working out personality/guild tone-wise.

    Meanwhile, the rational approach to finding a community would be exactly that: guild hopping. How else will you know if you get along with the people in the guild?

  7. Doone, I actually state explicitly, that some players love to go out and find new contacts. My life experience is that most people are not like this. And that's why I say it.

    Your life experience may be different and the people you interact with may be more extroverted than the people I interact with. That's fine. You are right that this doesn't much affect my argument.

    Can it be proven? Well, of course, you could start a scientific study. For a valid proof you would have to repeat that study whenever a new child is born on the planet ..

    This reminds me to make a post about this: "You can't prove this" assertion.

  8. Motstandet, thanks for the links. I read those before, but it was fun reading them again :)

    R. Koster is a smart guy. Unfortunately many of his examples are really dated by now. While this doesn't necessarily make them less relevant, it does make them less convincing at times.

  9. If the statements are defensible, by all means lets hear the defense. While you do state that some like to find friends/groups you have a quotable sentence above that says, flatly "I don't really enjoy joining communities and most people do not". Again, why the prevalence of guilds if people do not want to join communities? This is where I'm trying to take the discussion--I'm offering an opening to hear more of your thoughts on this, not shutting it down with "you cannot prove" statements. It's not about proof so much as a convincing argument to defend your points. And as I said before, I'm not convinced by your argument :)

    @Azuriel: Whether guild-hopping is bad or not isn't really what I'm getting at. Players are seeking communities is precisely what the bad trend of guild-hopping displays.

  10. Doone, I can't give you more than my experience to support the assertion that most people do not like the process of getting into a community and only enjoy being part of one (and the anticipation of being part of one).

    I realize that I might be wrong, but looking at my experiences with people, I consider it less probable than that I am right. There are no more arguments that I can bring forward. I don't know any studies on this and don't intend to start a search, because, in the end, it is not necessary for my overarching argument, as you already pointed out.

    You ask: "Why the prevalence of guilds if people don't want to join communities?"
    Because people like being part of a community, and if people are part of a community it has immensely positive effects for any product - especially an MMO.

    This does not contradict the statement that people don't like the process of joining a new community. For example, most people don't really like visiting the dentist. There are still dentists around, because people like having visited a dentist.

    Problem for the developers is that most new players in a MMO are not serious about the MMO. That's why they wouldn't anticipate the positive effects of being in a community and only dislike the process of becoming part of one. That's why you need to draw them (trick them) into a community. The best way, in my opinion, is the one I described under point (4) of this post.

  11. @Doone: I think Nils' assertion is supported by the whole LFD discussion: I've hardly ever seen people complain about group content based on the argument that playing with other people is not fun in principle, but people are still justifying the introduction of the LFD tool because finding people to play with (=joining a small community) was oh so tedious and hard work.

  12. While the some implementations advocated by Koster are dated, the fundamentals are still relevant--humans haven't changed subconscious socialization methods in only 10 years.

    The genre that Koster helped create doesn't really exist any more. Virtual worlds are gone, and MMORPGs are better termed Online Single Player Achievement Orgies. I'm not saying this type of game is bad (I certainly indulge from time to time), but they are completely different.

  13. I think what really would help with communites is social networking elements

    Similiar to how sites like work or netflix or amazon reviews. - it analyzes your preferences (based on your music libary, you last played songs and like/dislike) and groups you with similar players. I was amazed at the accuaracy, not only my neighborhood consisted of songs I generally like it also had groups and genres I didnt get exposed before but liked.

    So not only it correctly determines my currents tastes it able to predict what I will like .

    So allow players to put ratings on other players (like/dislike) which would be for purpose of matchmaking , analyze their activity patterns, and automatically group them. Making it more likely that you will be grouped with players from your neighborhood

    Thats is step 1. 2nd step is to integrate all the tools for functioning of a community into the game itself. Because the 2 tools which are necessary for a guild clan is forums/voice chat. And so far none of those are actually integrated in any games. That creates additional barriers on joining communities - you have to create external accounts and setup external applications which by default prevent forming the meaningful in-game communites.

    You cant voice chat with person you meet, you cant coordinate. So people default to interacting only with those in their teamspeak.

    Better social networking + all tools integrated inside the game = much better in game fostered communities

  14. In other words, this is impossible in a theme park ride that is called Wow, SWTOR and all their clones. The only possibility would be to grab a chair, smash the window, jump out and actually try to enjoy a fresh new virtual world. One can dream.

    Then again, maybe that's exactly what is missing. And the "we can't do it because players have learned something else" argument is total bullshit. By that argument humanity would have died out thousands of years ago.

  15. I like what Max proposes a lot. I'm not sure how we could go about getting a "rating" system in our games, but I agree that we can learn something from social networking sites on how to help players find others with similar tastes.

    Nils, it is true in my experience as well that players don't like the process. I guess I didn't understand your point at first. It seemed you were saying players don't like communities.

    I just recently made a post about the under-development of guilds in video games. I think you'll find it relevant to the discussion. In short, players are slow to change tried and true processes and game developers aren't really helping enhance this community tool.

  16. > You offer content that can be enjoyed alone. But
    > you also make sure that it is slightly more
    > efficient to group up.

    a) I don't think that works because I always try solo group content instead of grouping with other players, even if it's less efficient solo. I prefer to do something less efficient then to wait for the other persons afk breaks from time to time. ("afk baby" is just annoying if it's not your baby.)

    TO get me into a group it has to be "impossible" solo.

    b) Do we really want to group with people who like to do the same activity? Wouldn't bots be more useful for that?

    The meaningful interaction that happened in LFD is exactly the stuff that is not related to the dungeon run. Contrary to a) I might enjoy your story about your baby. But not when I try to finish some daily but when I have time for that.


    I think the most important step to draw people back into social communities for WoW would be to let player have more time and less preorganized daily tasks.

    Remove dailies, remove the daily dungeon, remove the heroic lockout, remove the VP/HP weekly cap. Remove everything that makes x amount of something the most efficient amount per day/week, because people will try to reach exactly x each day.

    After that, player will be willing to settle for less then x each day and again have time for each other. A wipe will no longer be an obstacle to reach your daily x but a chance to talk to each other.

  17. @Max: I have to strongly disagree with your point 2. Implementing voice chat (as an example for a community-fostering tool) is not a good idea in my opinion.

    There are three reasons for that:

    1) People who don't want to talk to other people they're playing with will not want to use it. That is different from people who don't want to type to communicate. Talking can easily remove all immersion from a game. Besides, I'm not sure whether you have played any WoW in the last 4 years, but, can you remember anybody ever using WoW's built-in voice chat feature?

    2) People who want to talk to other people generally have other means available to do so anyway. All guilds I've ever been in or heard of had at least a ventrilo or teamspeak server. That also has the distinct advantage that you can talk to people while they are not logged in, or try out other games.

    3) Considering 1) and 2), it is questionable whether the marginal improvement is worth the amount of development time you'll need to invest as a company. I still remembers with pain WoW's patch 2.2, which had not much of interest except a voice chat feature that was never used, and took as long as any major content patch of the time to finally get to release.

  18. I am guilty of not reading the majority of comments here -- sorry, maybe something like this was already mentioned.

    Anyways, Nils, what do you think (in context of the title's post) of what DDO was doing (maybe is still doing, I don't know) -- basically the majority of content requires a group AND there's a grouping tool. Only instead of being anonymous, it is very much a "full disclosure" deal. Someone has to create a group 'ad' which appears in the game's grouping interface. It is possibly to specify the specific instance you want to do or simply put any generic comments (e.g. I've seen something like "first time players -- no spoilers", etc.). Then people can send requests to join the said group.

    Do you think this works to make a better community?

    P.S. Blogger comment input sucks big time.

  19. I haven't played DDO, Solf, so take my opinion with a grain of salt.

    Considering your description and what I know about DDO that group finding tool fits perfectly to the game. Even in a Wowlike I would prefer this system to the LFD. A complete analysis of this system doesn't fit into the comments ;)

  20. As a kind of follow-up to this -- Nils, if you're interested -- do you think you could post your opinion as to why WoW went the way it did with grouping tool?

    I mean the original one was kind of sort of more open about what it did... it just sucked very much from the interface point of view (imo)... why'd they not improve that and just jump off the cliff into DF?

  21. I might blog on that in the future, Solf. My current explanation is that Blizzard's devs are hardcore raiders who try to appeal to the casuals they see in their metrics.

    But being who they are and being part of estabilished communities themselves, they are unable to design for casuals with their 'heart', and have to rely on metrics to an extend that that is not always good for the game.

    But that's certainly not the final explanation.

  22. Ah that one... I didn't comment back then, but your theory didn't make sense to me.

    If that was done for/by the hardcore, where's MgT in WotLK and Cataclysm? (just an example)

  23. Magister's Terrace? I don't think I understand.

  24. I think Halls of Reflection heroic was more challenging then Magisters' Terrace heroic (both on release).

  25. Yes, Magister's Terrace. A seriously challenging content (imo) for even the raid-geared players. Hell, where are actually (seriously?) difficult heroic modes in 5 mans?

    Kring says that Halls of Reflection might have been it. I don't know -- maybe it was. I have quit WotLK before that out of sheer boredom (no challenging content at all aside from raiding).

  26. My stance on the matter is this:

    You can't have seriously challenging content, a random LFD and a daily quest. This combination spells disaster. Even if you remove the daily quest the combination doesn't work. Blizzard discovered that (hopefully) with the first Catacylsm heroics, that were sometimes extremely hard in the beginning.

    The only way we will get challenging non-raiding content back is if this content is not accessible by LFD. You will notice that when the LFR is released: The raids will either be very, very easy or the feature will be a complete failure.

    In MoP the only challenging content left will be normal and heroic raids. And they are exactly the content that is not accessible via LFD/R, but requires socializing with your server community.

    Since PvE Scenarios will be accessible by LFD they are not the solution they could be.

  27. I agree with your assertion that 'difficult' cannot co-exist with DF.

    This still doesn't answer the question of how you came to original conclusion that "WoW was dumbed down for the hardcore" in the face of the fact that there's exactly zero challenging content except for the raids.

    Hell, it'd take really _hardcore_ to raid every day :)

  28. It is the very definition of a hardcore raider, Solf. I wrote:

    "They just want heroic raids and high-end PvP. They half-heartedly dumb down arbitrary content trying to fulfill their obligations. "

    I'm sure the hardcore raiders would also like challening 5-mans. But they sacrificed this to the LFD, assuming that there is no way the two can co-exist. They are wrong. You can have some dungeons that are accessed by LFD and others that are not. Call them scenarios if you want.