My Simple Minded World

A Framework for Understanding Motivations

Posted in business, The Interweb by Omar Ismail on April 16, 2009

Audience. Community. Social. Viral.

These are all buzzwords that have gained popularity over the past few years to try and capture the fuel that has propelled success stories like Facebook and YouTube. But what does this actually mean, and how can we recreate them?

Based upon the work of Amy Jo Kim (Building Communities on the Web), Richard Bartle (Players who suit MUDs) and Andrew Chen (Futuristic Play) I’ve developed an analytical framework for evaluating ideas and initiatives on how they impact a person’s attachment to your site.

Motivational Taxonomy


I split up motivations into two broad categories, “External” which I qualify as things that will affect a person’s real life in some way, and “On Site” which is limited to things on the site itself. While the focus of this article is on the “On Site” motivations that isn’t to say that “External” motivations aren’t important. In fact I would argue that a large part of Facebook and Twitter’s success come from how the services affect a person’s real life. However, these External things are pretty self-explanatory and don’t need to be covered in depth. In my research I found the OnSite factors to be a more difficult thing to quantify and that’s what I’m attempting to model.

Identity (I)


This refers to a person expressing themselves to the world around them.

Who am I?

What am I about?

What do I like?

What do I want to do?


As a benchmark to illustrate what I mean by Identity: Facebook. People’s profiles are an incredibly close approximation to who they are as a person as people use their real names, real pictures, real interests, real friends.

MySpace is another obvious example but in a different way as the profile pages are focused on manufactured identies and creative customization.


Relationships (R)


This motivation is defined by a person being socially engaged with another person or persons. Humans are social creatures and we want to have a sense of belonging.

We want to communicate with like-minded individuals, have shared experiences, and shared traditions.

A benchmark for relationship building is MySpace in the old days, and Twitter now. These services are about building a broader and broader breadth of relationships. Some services like niche social networks, or hardcore MMO games focus on increasing the depth of people’s relationships to one another.


Content (C)


This is the person’s affinity towards the actual content on the site. They may be passionate about a particular topic, enjoy the nature of the conversations, find the content funny/entertaining, etc. We all have interests and content that we like to consume, I like watching Street Fighter 4 videos instead of cooking movies.

A benchmark for content is Wikipedia. For any topic you might have an interest in, they have a wealth of high quality content to consume.

I would argue that for outsiders, chatrooms rank low on the content scale because the individual conversations present little value.


Weightings and Score

Every site/service can be graded on these three metrics and scored relative to whatever benchmarks you choose.

This isn’t the entire story though, as each individual has their own importance and weighting they apply to each metric.  And a person’s weightings will shift depending on the topic at hand.

These weightings combine with a site’s metric rankings to get a person’s individual site score.

SCORE = w1*I + w2*R + w3*C

The resulting SCORE determines how much attention a person will give a site. As long as the score falls above that threshold the more time/attention a person will give. Therefore stickiness is achieved when (I,R,C) increases the more time a person puts in.


Applying it to People

This is all well and good, but how do we apply this to people, and how do we know what people’s weightings are? Well this is where Bartle’s paper comes into play. From analyzing MUDs in the 90s he proposes that there are 4 distinct classes of users/members.

  • Achievers – get satisfaction from overcoming obstacles, gaining ranks, and improving their status
    • C >= I > R
    • Sees the service as a game (chess, checkers)
  • Socializers – want to talk and build relationships
    • R >> I > C
    • Sees the service as entertainment (bar, club, etc)
  • Explorers – want to see/experience everything the system offers and testing the boundaries
    • C >> I > R
    • Sees the service as a pasttime (gardening, reading)
  • Instigators – enjoy stirring up trouble
    • I >= C > R
    • sees the service as a sport (hunting, fishing)

So we can attract or at least make our sites more attractive to these personality types by creating features, policies and intiatives that appeal to them.

What makes things really nice and messy is that each of the people in the system has a substantial effect on one another. Bartle provides a nice interaction diagram (I call Instigators what he calls Killers).


So you can see that the more Killers/Instigators there are the fewer Socializers. More Achievers and Socializers brings more Instigators, and Explorers are in their own world.


Bringing it all Together

This analysis brings a few questions up

  1. What kind of population distribution makes the most sense for your site/service?
  2. What initiatives do you do to appeal to each personality type in the context of your site/service given you better understand their motivations?

While these questions are still difficult to answer and require creativity/innovation they’re at least a lot more grounded and workable than generic statements than the usual “let’s make the service better”.