Personas
A technique that employs fictitious users to guide decision making regarding features, interactions, and aesthetics.
The design that seeks to accommodate everybody generally accommodates nobody well. For example, the percentage of visitors who actually buy products on an e-commerce website is typically quite small relative to the total number of visitors, but most website designs (and redesigns) fail to consider the differing needs of buyers versus browsers — that is, they design for the average visitor, an impersonal and homogenized construct derived from sources such as visitation statistics, surveys, and usability testing. It is better to understand and perfectly meet the needs of the critical few than to poorly meet the needs of many. It is this specific problem that personas seek to address.1
Personas involve the creation of profiles for a small number of archetypal users, each profile representing a composite of a subpopulation of users. Information for the profiles is derived from user and stakeholder interviews, reviews of market research and customer feedback, and statistics about how a product is used when available. Done right, the number of personas is small, with typically no more than three primary personas representing the primary target audience, and up to four secondary personas when the needs of the user population are highly stratified. Each persona is typically represented with a photograph, name, description, and details about specific interests and relevant behaviors. It is often useful for members of the design, development, and testing teams to role-play different personas. This clarifies user needs and behaviors and is an effective means of creating empathy for the user perspective. Personas not only make the target audience more real to designers and engineers, they also ensure that requirements are prioritized to specifically meet the needs of high-value users.
The use of personas in the design process is increasing in popularity, though empirical evidence for the technique as compared to more traditional approaches is lacking. The measurable merits of the approach are difficult to ascertain due to the proprietary and relatively secretive nature of the methodology developed by Cooper. The unfortunate result is an abundance of teachers, consultants, and practitioners engaging in their own version or interpretation of personas. Nevertheless, the anecdotal evidence for the general approach — especially its user-sensitizing impact on designers and developers — is compelling.
Consider personas early in the design process to define and prioritize requirements. Keep persona profiles short, preferably one eye span, so that the information can be easily consulted. Limit the number of primary personas to three and secondary personas to four. Base personas on interviews and market research — do not make them up or recycle personas from past projects. The time required to research and develop personas is generally less than one month.
See also 80/20 Rule, Desire Line, Iteration, and Normal Distribution.