I often hear people talk about personas. Although they correctly attribute them to Alan Cooper, the descriptions I hear often conflict with what I have read in Cooper's books and learned at his companies training seminars. In this post I will try to point out what I think are some important differences between what I have learned and what I often hear described.
To try to keep things clear, I will refer to personas as I hear the typically described as Typical Personas (TP) and personas ala Cooper as Cooper personas (CP).
Discussions of personas are usually related to the requirements gathering phase of a software development project (Cooper does not limit personas to software, but I will for this discussion). The first step in this process is generally to determine the users for the proposed software. The typical process that I hear for generating TPs is described like so: come up with a list of users (usually by brainstorming), consolidate this list and organize into roles, then if you have time and feel it is necessary (would be beneficial), you can create personas by adding some interesting personal details and a picture. While this is a crude description, it fairly describes several conversations I have heard; the common points being 1) personas are created at the end of the process 2) as an extension of user roles 3) they are created by adding personal details to create an imaginary user that project members can identify with.
A CP is a model that is based upon observed behavior, that is, research. User roles are identified to help determined research sample sizes. People trained (special skills) in interviewing and observation to field research to gather information about the users behaviors and environment related to performing business tasks. This may include using software while performing those tasks. Observed behaviors, motivations and goals are analyzed and used to identify personas. Narrative descriptions are developed for each persona.
Each process appears to produce a similar product (a narrative description of a user), but it is the process that leads to each that is important.
TPs tend to start with a brainstorming of potential users. CPs start with observed behavior. In my opinion, this is one of the most important differences. Starting with direct observations by people trained in observation and interviewing techniques is far superior to brainstorming exercises, postulating, or remembering past activities.
TPs tend to have a 1 to 1 relation with a user role, after redundant roles are eliminated. CPs are modeled after patterns of behaviors and goals; they typically represent the needs of many.
I believe personas are a powerful tool for product design and I hope this post will encourage others to further investigate personas further and learn how to use them in their development processes.