I recently sat down with one of my colleagues from Knowledge Management who has been working in the area of customer feedback for quite some time. She really has a good feel for how to analyze and approach results, as well as set up a project from the get-go.
She and I were speaking about the definition of the audience and the roles of the people interviewed. She found that defining the roles and responsibilities of the target audience, in particular, what documentation they access to execute their work, gave real meaning to her results. She broke down the roles into technical and process experts, as well as their relative levels of experience in those roles. She then created a personae for each task-based role that a person performs at the company.
In addition the questions were generic enough that they could be answered by all the different roles and people being surveyed. You can’t always drill down as much as you’d like, but this was still a good approach for the team’s needs.
From these almost real personae, she could derive useful generalizations based on “Bob the Programmer” or “Steve the Consultant” or “Miriam the Support Specialist” or whatever she and her group of feedback coordinators deemed appropriate.
But her results didn’t lead to generalizations that were too wide. She did not say “all developers found this documentation useful.” She took samples from both ends of the spectrum to highlight what went well and what didn’t and what was accessed and what wasn’t. The persona was really the starting point so that people looking at the data could give it a human face and put it in a context of a real working environment.
The only real danger with using roles and personae is that there might be preconceived notions of what the role and tasks might entail or if that role no longer exists at the company (which does actually happen).
However, this is an approach I can heartily recommend because it gives you good results and a good story to sell to your manager when you are asked “SO what did you learn?”