Initially, we took the results of our five interviews and went through them, making a map of each concept and the answers that we collected. Our idea was that this would help us to decide which things our prospective users had in common, and which things were unique. We used several large pieces of paper to map this out.
Our group developed personas and goals for the system based upon our interviews. Our process was as follows: we iterated jointly a list of questions for a qualitative interview process, and then conducted the interviews with a wide range of people who reflected people we thought fit the target users. These users include those who are not at all aware of or used to blogs, to those who read heavily, to those who were writers and creators. Interviewees reflected their reading, writing and search habits, and their familiarity with technologies that support these habits both offline and online. The following is a summary of our findings:
Information Sources: Ultimately, the reliability and relevance of news sources were the factors that led each of our interviewees towards their preferred news sources. This put sources like The Economist and The New York Times at the top of the list. However, people found referrals from friends and referrals from blogs to be nearly as important, simply because it helped the person to select the most relevant news.
Information Seeking: There were three major behaviors that our interviewers explained to us. When in need of information, they would use a search engine, ask another person, or go to a source that they considered to be authoritative on their topic. Each of the stories the interviewees told us involved some combination of those actions. Interestingly, in a majority of the stories, the person started with Google, but ended up speaking with or e-mailing another person for the information.
Reading habits: Overwhelmingly, the people we interviewed loved to read. However, many did not have the time to read as much as they would like. These people found ways to maximize the signal-to-noise ratio in the information they read. They would use social filtering (recommended by friends) or trusted source filtering (reading from a limited number of sources with which they had good experiences).
Writing habits: We found that people either loved or hated writing. Those who loved writing did it almost every day and considered it to be a craft. Those who disliked writing felt that it was a painful or difficult process. Another enjoyed writing, but felt that it was a lot of work because he insisted on perfection. Some wrote for themselves, others only wrote if they knew they would have an audience.
Blogging habits: There was a wide variety of experience and familiarity with blogs in our set of interviews. All had heard of blogs, yet some did not have an accurate idea of what a blog was. Others were so-called “A-List” bloggers who maintained their own blogs and posted on a regular basis. In the middle, there was the set of people who read blogs as a method of information filtering. These people either maintained blogs for occasional posting or did not keep a blog.
Perceptions of blogs: This question had an incredibly wide range of answers. Some people felt that blogs were simply narcissistic outlets for the vain. Others felt that blogs were a kind of new media source or an alternative to traditional media outlets.
Blog writers: We found out that some bloggers are fixated on knowing their ranking relative to other bloggers (based on the number of references made to their blog). Other bloggers had never even looked at the number of hits their site was getting or were specifically not interested in knowing this information.
Advertising: In the interview with the market development manager, we found out that some companies had PR departments who monitor online chatter (ie, rumor web sites). We also found that advertising and marketing is very dependant on quanitativequantitative results and each company has a different philosophy as to what an effective sults and the .
We then used the interview feedback to determine a list of activities and needs, which then led to generating the underlying goals that drove the activities people described. Tasks were then determined from the list of goals and activities. Upon seeing what those tasks and goals were, we then constructed a range (approximately 12) of personas, documenting their goals, lives and tasks they perform.
After revising the goals, as we discussed the possible uses of the system and the reasons why our users would want to do certain things, we chose four personas and mapped them across the task list, to see how important and how often we thought those users would do the tasks.
Our group met in person to document and share our interview findings, discuss our understandings of these issues, and determine what our personas and goals would be as mapped from the interview results. We then worked separately, dividing tasks and sharing the resulting documentation via email. We met again to talk through the personas and tasks, to make the most accurate and honest representations we could achieve, based on our feedback and understanding from our user interviews.