« Size Zounts | Main | Cyber Erotica and the Disappearing Arts »

Jul 24, 2006



On two recent blog posts (shameless plug alert) at www.tinkerx.com, I proposed two new terms: "social share" and "share of participation" to try to define, first; (social share) how much of a given participatory market a particular company, service, product, etc., has, and, second (share of participation), how much of an individual's attention and "care" is given to various social activities.

A guild wishing to maximize its "social share" (measured by number of members within a shard, quality of members, time spent by members, quality actions contributed by members) would be looking to do one of three things. A) Increase the gross number of members participating in the guild and contributing at a low level (high volume, low participation per member). B) Increase the "share of participation" of each member (raise the quality of interaction of each member). C) Both.

The decision about when to "half-life" (don't you love English? I can use "half-life" as a verb...) an inactive guild member would (should) be made based on which of the above marketing strategies was being utilized by the guild. If you were concentrating on A. (high volume), you'd want to leave a long half-life period, because big numbers would be key to your strategy. Even if somebody comes in for 10 minutes after a 3 week absence, if you've got other activities, tactics in place that revolve around the "Look how many dudes we've got!" message, you don't want to foul that by kicking people out. If, on the other hand, you want to encourage high "share of participation," (soon I will start leaving off the quotes, my friends...) you will want to encourage a certain level of "if no X activities, no Y rewards," and perhaps, "if no activity at all, then punishments," including the big, black boot itself.

I signed up for "Puzzle Pirates" about 3 months ago just to goof around. I don't like puzzles that much, but I was curious. I played for about 2 hours and thought I'd probably play for about 2 hours a month. After 2 weeks, though (I think, wasn't paying attention), I got an email from them saying, "Hey! You haven't played in awhile. If you don't log on and play soon, we'll give you the big black boot."

Well... I did log on, because I'm still curious, from a research perspective. But from a pure gaming/gamers perspective, my gut response was, "F*$k you, Mr. Puzzle Pirate King!" That's like Cinemark calling me and saying, "You haven't been to a movie in 3 weeks... If you don't come to see a flick soon, we won't let you back in!" Feh. They may be trying to incent a higher level of share of participation than is warranted for a puzzle-based social gaming service... or they may be trying to get rid of the 2-hours a month gamer. Which would be me. If so, then they are right on target.


The half-life of a guild member is pretty much dependent on the guild I think. In the guild I currently play with, there seems to be no half-life at all... members in good standing seem essentially immortal. They can die off for years (or even game-eras... leaving during EQ and returning for WoW after skipping DAoC and CoH) and return to good feeling and excitement, which is in my opinion how it should be.

Other guilds are much more strict, and that's the way they want it, or at least believe it should be. I think it probably has more to do with the play style and personality of the guild's core people tan it does with anything inherent to the guild structure or MMO design itself.

A problem with MMOs ability to retain customer/players however is the inability for someone to take time off and return later. I took some time off after hitting max level in WoW and returned a few months later to find myself pretty behind. Luckily my guild is pretty helpful and understanding, and after I got the prerequisite gear upgrades done (fire resist in this case) I'm playing with the majority in MC and on Thursday we're going into BWL. But a RL friend is in my same position I was in, and it's just not worth it to him to "catch up" to the other 60s, so his account will probably be canceled sooner than later.


A point of clarification, Nate. By "half-life" of an inactive guild member, are you referring to the amount of time they're inactive until a guild leader boots them?


By "half-life" of an inactive guild member, are you referring to the amount of time they're inactive until a guild leader boots them




Nate, if that's how you're defining half-life, I do believe you've stretched the term beyond recognition.

The question you raise is interesting, but I don't see how it relates to the seed article. =]

And doesn anyone here actually believe that by studying a Hungarian news site people are gaining insight not only into the rest of the Web but biological systems as well?

Scale-free network theory applied like this is hocum.



Nate, if that's how you're defining half-life, I do believe you've stretched the term beyond recognition.


Oh okay, would you believe 2 * 1/2-life = guild boot? Geesh.

The question you raise is interesting, but I don't see how it relates to the seed article. =]

And doesn anyone here actually believe that by studying a Hungarian news site people are gaining insight not only into the rest of the Web but biological systems as well?

Scale-free network theory applied like this is hocum.

As for the analogy of online-id's and Hungarian news stories. Poetic license. However, it is amazing sometimes (to the stories I have heard) how forgettable most of us are online if haven't logged on for a while. In the first case 15 minutes stretches to 36 hours, in the second...?




Interesting topic and question. Few comments:

1. While the average life of a news article may be 36 hours and follows a power law, the more valuable knowledge is how to extend or shorten the time frame. There are probably much more research on this area.

2. "What's the half-life of an inactive guild member..." is an interesting question. It's a cost-benefit decision that guild leaders make every day. Again, it's would be good to see data on the average and median time. We may even find something like a Dunbar Number.

3. Interesting to hear what are the average half-life for other membership networks or even deliquency periods. I'm guessing 12 months for memberships and 3 months for deliquencies.

4. There is a conceptual trend in business of a) working hard to keep your top 20% clients and b) working hard to disengage with your bottem 20%. What is the average half-life for the bottem 20%? What is your guess?




In my WoW guild, we kick members after 30 days of inactivity. So, if someone doesn't log onto their character at all for a month, they're gone. I suppose we're focusing on the "high share of participation" that Andy mentioned above. We have about 150 people in our guild, with a core raiding group of about 50. Of that core raiding group, approximately 15 are officers. We encourage high levels of participation, as an endgame raiding guild, and provide incentives for more frequent attendees (in the form of raid invite and loot priorities). From what I've seen of guilds on my server (Hyjal), the 30-day boot is pretty standard.


@Rae: YEA! Someone else used my term! I'm so happy... Thanks.

There's a term in some marketing/group-building circles I've heard called EIEUEO; "easy in, easy up, easy out" that refers to how you should go about structuring rules for the development of a large, but not particularly "deep" or long-lived type of audience or customer base.

"Easy in" = making it very simple for people to initially join your group or get your product. Low initial price point, no or low equipment cost, no credit check, no minimum age requirement, no complicated learning curves, etc. This is the "fat pipeline" that lets tons of people in the front door, no holds barred. Or no bars held.

"Easy up" = easy ways (hopefully more than one) to advance to more interesting, complex or deeper types of relationships. If I start out, for example subscribing to a free email newsletter that only requires I supply an email address, I can then "move up" to 1) a free monthly "deeper content" e-magazine that requies me to fill out some demo info; 2) a minimal subscription-cost paper mag; 3) a blog/portal where I can spend some time and contribute to the group; 4) a listserv; 5) a website where I can buy swag. Any of those options represent an "easy" choice after having made an initial "easy in" decision. And having multiple choices makes them all easier, because it doesn't force me to make just one. Many retail studies have shown that having choices makes people more comfortable making ANY choice.

"Easy out" = both a way for the member to leave without penalty (so that they don't feel bad about it and will come back if they so choose later), AND... an easy way for the system/group to boot out folks based on easy rules of conduct. If you are trying to build a kid-friendly space, for example, a "no R-rated language, one warning and then you're out" rule is an "easy out" clause. And it should be followed to the letter, or the folks who stay in will be less likely to stay.

This will tend to build a "wide" group but, as I said, not a really hard-core, vertically homogenous one. "Easy in" alone is enough to trouble some group management folks, as many systems "want" more rigorous requirements for entry. Which is fine. Some groups require fewer, more specific members.

But... ask yourself this when you put a 30 day "No activity" limit on your guild members: could you have an "easy down" rather than an "easy out" for them? Is there a stage of discipline rather than all-out "buh-bye" that would make sense? Because the "easy up" policy is one of the ways that truly builds customer loyalty in business; and giving "fallen" guild members a chance to earn back their "up" could provide you, eventually, with your most die-hard players.


Hm, that's a very interesting perspective, Andy! We actually do demote them to a "holding pattern" rank after 25 days of inactivity, and if they come back before the 30-day boot, we permit them to earn their way back up. We also do make exceptions for people who inform us beforehand of extended absences -- we have many guildies who are active members of the military, for example, and we would never dream of kicking them out during a deployment.

That said, what we really value in our guild is communication and respect. Disappearing for an extended period of time with no notice shows neither of those things, and in the end, we don't want to hang out with people who don't possess those qualities.

Back to your perspective on marketing strategies, I would have to say that it depends strongly on how incentivized people are to join a group. In an environment such as World of Warcraft, there are guilds who are loose confederations of people that would be very well-served indeed by adopting the EIEUEO strategy for recruitment. It would provide a constant influx of people to keep the guild vibrant and motivated. However, there are also high-end raiding guilds which, because of the high volume of applicants, can afford to be much choosier, and impose stricter requirements/rules on their members. So I would imagine that the half-lives of members would vary wildly depending on the type of guild one was looking at!

The comments to this entry are closed.