The Problems of Using Personas
Have you been using personas in your design process? Below are some issues to take note.
The concept of using personas originated from Cooper (1999), who observed that designers were often unclear of user needs. In response, he proposed the use of personas — fictitious human characters with attributes derived from ethnographic user research. Pruitt and Adlin (2006) identified three benefits of using personas: helping designers shift their focus from self-centredness to user-centredness; promoting better understanding of users’ complex and diverse needs; acting as a proxy for situations where users were inaccessible. Matthews etc. (2012) observed that personas were mostly used to champion user needs, or garner support for a chosen design; a persona might also be used if time or resources did not permit them to gather such first-hand experience, even if this option was viewed by many as sub-optimal. Pruit and Grudin (2003) argued that the use of personas, in conjunction with other techniques, could restore the social and political dimensions of the connection between designers and users.
However there were many problems with using personas.
Firstly, Matthews etc. (2012) encountered designers, developers and others in design teams who had difficulty “believing” in personas — how “all 40-something women shoppers in Mid-West” could be turned into one persona, “Katie”; they found personas to be abstract generalisations. Chapman & Milham (2006) stated that any persona would represent a small proportion of a population only. It was difficult to ascertain how representative a chosen set of personas would be, and how important the users being represented were to the design problem.
In addition, Chapman & Milham (2006) explained that as more personifying details were added, the persona became less representative of the population instead. Each characteristic embodied less than 100% of the population — a probability of less than 1.0. A persona comprising of a composite of characteristics meant the more characteristics it had, the lower its overall probability (population representation) would be. Moreover the overall probability would be affected by the multivariate correlations of all of its characteristics — a highly challenging and exorbitant set of data to determine. Chapman & Milham (2006) also commented that such correlation data had never been determined for any persona, which implied the impossibility of assessing a persona’s accuracy or significance.
Secondly, Matthews etc. (2012) realized that the personas were impersonal — the personifying elements were unable to evoke the same sense of empathy in designers that first-hand observations of real users could. Designers often had to draw upon their own experiences to imagine the persona’s behavior, which was a sign of information gaps left unfulfilled by the persona’s personifying details and narratives.
Thirdly, there was insufficient information provided by personas that allowed designers full understanding of the users (Matthews etc., 2012); the personifying details (e.g. interests, personal habits) of the personas contained information that were supposed to build empathy and assist designers to anticipate user actions and emotions, but these details also included information that could be irrelevant. These would result in the formation of false constraints that mislead designers instead. In addition, Chapman and Milham (2006) highlighted that since personas were fictional, no data could invalidate an imaginary construct; this, together with the lack of verifiable data in the persona literature, made validation of constraints challenging and claims of successful use of personas dubious.
Lastly, the irrelevant personifying elements were distracting, sidetracking designers from constraints that mattered (Matthews etc. 2012). More unique aspects of a persona might become more obvious because it was easier to connect with them. Eventually, the personas created would then make up of “nice looking pictures and fancy job titles and extremely exciting lives”, at the expense of more important attributes.
Making Better Use of Personas
Blomquist & Arvola (2002) advised that in a project team, there should be someone who was familiar with creating peronas. The experienced person would then be able to guide the rest in the creation of personas and scenarios. To make the designers trust the primary persona, all designers should be involved in the persona creation process, so that all designers understood which parts of the personas were assumptions and which were built on empirical material. Participating in the persona creation process would also help designers attain a richer understanding of user needs, so that they could then use the personas and scenarios created as communication tools to articulate user needs to other stakeholders.
To furthur mitigate the risks of using personas, Matthews etc. (2012) advised that personas should not become a substitute for spending time on user observations. Designers should immerse in the real environment to obtain first-hand user observations or access to user study data. Doing so was essential to complement the use of personas, so that designers could obtain a holistic understanding of user needs, and better understand how user data was synthesized to form composite personas.
During my Hyper Island journey thus far, all of my project teams did not employ the use of personas. We collected quantitative data through online surveys; qualitative research was conducted via on-site user observations, random street interviews, and Skype or face-to-face in-depth interviews. Literature review was performed via academic journals and online industry publications, acting as secondary sources complementing our primary research. We were always able to access users and had sufficient time to engage users for primary research. For the Business Strategy project, my team also managed to invite users to participate in co-creation.
Based on client and industry feedback, our arguments constructed were sound and persuasive, and none mentioned that the use of personas would have improved our understanding or communication of user needs to others. Therefore I learnt that although designers commonly used personas, I should not blindly adopt the use of personas. Primary research was more effective than using personas to imagine what user behaviors might have been.
Blomquist, Å. & Arvola, M. 2002, “Personas in Action: Ethnography in an Interaction Design Team”, Proceedings of the Second Nordic Conference on Human-computer InteractionACM, New York, NY, USA, pp. 197.
Chapman, C.N. & Milham, R.P. 2006, “The Personas’ New Clothes: Methodological and Practical Arguments against a Popular Method”, Proceedings of the Human Factors and Ergonomics Society Annual Meeting, vol. 50, no. 5, pp. 634–636.
Cooper, A. 1999, The inmates are running the asylum: Why high-tech products drive us crazy and how to restore the sanity, Sams Publishers.
Matthews, T., Judge, T. & Whittaker, S. 2012, “How Do Designers and User Experience Professionals Actually Perceive and Use Personas?”, Proceedings of the SIGCHI Conference on Human Factors in Computing SystemsACM, New York, NY, USA, pp. 1219.
Pruitt, J. & Adlin, T. 2006, The persona lifecycle: Keeping people in mind throughout the product design. Morgan Kaufman.
Pruitt, J. & Grudin, J. 2003, “Personas: Practice and Theory”, Proceedings of the 2003 Conference on Designing for User ExperiencesACM, New York, NY, USA, pp. 1.