Personas Are Still a Mystery
August 23rd, 2005
Adaptive Path’s Dan Saffer recently wrote an interesting article on Personas. Dan has observed that many development teams build personas without conducting up-front user research.
I’m not really surprised by Dan’s experience. Clients come to us all the time wanting to know if they can build them without doing any research.
Lately, in the hope of better answering our clients questions about personas, we’ve been looking closer at how development teams use personas and how they incorporate them into their design process.
One of the biggest surprises from our research is how each development team leverages personas in a slightly different way. While most of the teams were aware of the Cooper technique for developing and refining personas, very few adhered to Cooper’s rigorous process for building personas from the user data. It wasn’t that the teams didn’t want to use Cooper’s approach. In most cases, they just didn’t have enough information about how Cooper actually builds the personas. The detailed process was a mystery to them.
Most of the teams we investigated did conduct up-front research to collect information about their users. Some teams created fictitious personas based on interview and focus group data, while others based their personas on actual people they had interviewed. We also found that, when teams use personas well, every member of the team really does seem to be on the same page about who the users are and what design will work best for them.
While our research is still underway, we’ve been so happy with our preliminary findings that we’re going to continue to take a closer look at what’s really happening when a team uses personas. We’d really like to take some of the mystery out of persona development.
Are you using personas in your development process? Are your personas based on user research? What benefits are you seeing?
August 25th, 2005 at 2:40 pm
As well as with scenarios there are three classifications: actual, factual, and fictional. Each can serve a purpose and is generally ‘dictated’ by circumstances. Conducting “actual” research is not always an option. In other cases (proposals, for example), a ‘factual’ personna can be developed by generous research about a particular ‘type’ of individual and their motivations. Not applicable to ‘general’ audiences, in one case I was able to do considerable research about the flight attendant industry and what might motivate a person to work in a career where the average starting salary moved less than 1% in 10 years (government documents online) and typically was no more than $13,000 annually.
As more ‘facts’ become available new personas are added and others modified.
From my observations, not including personas is a cultural issue. Many technical floors are decidedly ‘old guard’. Even the requirments people are leery of them. There isn’t anywhere that these ‘new’ deliverables fit into their highly prescribed methods of operating. The operating cultures we created in the dotcom era were decidedly rogue…classic operating methods were tossed out the door. We defined the ‘details’ in composite methods. To strip apart the detail inherent in a persona and represent it in details of ‘facts’ could turn a 1-2 page representation into 50+ pages.
The persona is a icon of the clash between trusted interpretation and ‘hard facts’ (the latter being an anomaly of reality anyway…but people believe in them).