Tuesday, November 1, 2011

Playerbase and Userload

When it comes to growing a playerbase, I've become more convinced than ever that it's got to be a staged process unless you have a LOT of money to blow on it. Having a lot of money is in fact how a lot of big MMOs handle growth - when they officially open, they basically go to full capacity overnight. This means that they can focus all of their development effort on one specific expected player load, and don't have to worry about nursing a small user base through the various stages to become a large user base.

This whole staged growth process has become a lot more apparent to me over the last couple of years:

  • When the player base is very small, it attracts a certain class of isolationist explorer player, one that doesn't need much in terms of a help system, documentation, or maps.

  • A bit larger and you start to get power levelers that like to compete with each other for in-game prestige.

  • Larger still and you start to get quasi-social players, players who like to group with others and help out. They don't show up earlier because explorers don't like to group, but with power levelers there's someone to group with and someone to help. This stage is an easy place to stall at, because these quasi-social players do not have the thickest hides and are less competitive than the power levelers; if they are abused by the power levelers, they will simply leave.

    Power levelers tend to form small cliques with these quasi-social players, and then drive off all other quasi-social players not in their clique. This appears to happen because the power levelers in one clique, competing against those in another clique, try to get an advantage by driving off the other clique's support staff.

    If restrictions, rules, and code are not put in place to protect them, the quasi-social players will never reach critical mass and will cycle out as fast as they cycle in.

  • The next stage brings in true socialites, players that typically gain levels only as a side effect of being in a group that is running experience while they chat.

    True socialites are another huge rats nest of problems. They tend to be drama queens or drama magnets, and are as capable of stirring up political *CENSORED*storms as they are of keeping other players interested and happy. There isn't a lot that can be done about this, other than to shutdown flamewars and ensure that rumors about admin 'unfairness' are killed immediately.

    True socialites are heavily looked down upon by the explorer and power leveler communities, and they must be given substantial protection from abuse. Unfortunately, socialites can tolerate a lot of abuse - unfortunate because when they take abuse, they dish it back out tenfold in the form of bitching to other players. One sufficiently pissed off socialite that isn't immediately shut down can crater the happiness and opinions of two dozen other players.

  • The last stage that I have familiarity with is the 'true newbie' stage. This is where you get true newbies coming in, and your game has progressed far enough that true newbies are actually able to make some forward progress.

    True newbies are a huge problem. From a technical standpoint, they are generally completely ignorant of the game style, and the newbie areas must be very well built to retain even a fraction of them. This category also appears to contain the largest percentage of 'low functioning' individuals - I do not mean that as an insult, rather an observation. Not everyone has the mental capacity to be an explorer or a power leveler.

    These new players must be afforded additional protections that even the socialites did not require. I'm still working out the details of a lot of this, but a good chunk of it amounts to protecting these players from themselves.

These are just the social problems. There are also scaling issues, each of which presents a 'critical mass' hurdle that must be passed. Certain areas may require a minimum number of players to be handled; assembling a group of that many players may be impossible below a certain user load. In the case of Alter Aeon, many of the automatic control systems rely on statistical data to function correctly. With few players, the lag time on getting statistically relevant data may be literally months.

This kind of lag time really hinders the ability of a game to make quick and effective updates to increase retention. If you have a thousand new character creations per day, you can pretty quickly determine if a change is an improvement or not. If you have one new character creation per day, it becomes very difficult to determine if you're doing the right thing. I suspect this is at least part of the reason that MMOs take out funding in order to start big and improve the game quickly.

I think at Alter Aeon, we're past a lot of these initial hurdles, and could probably reboot the game from scratch pretty quickly; once you get the various protections in place, they remain largely functional even though they fall into disrepair at higher userloads.

That said, I know there are bigger, nastier hurdles ahead of us.  I just don't know what they are (yet.)

No comments: