I am not into blogging at all, as is evident from the fact that I am the last in EDI to paste my blog link. But finally I am here giving blogging a shot and lets see if its as addictive as many people claim it to be.
Since we are into the research phase of our class design project, talking about the troubles in interpreting user research was the first thought that came into my mind while thinking of a blog topic. User centered design (UCD) makes a lot of sense in design of products which not only fulfill customers' needs but do so in a seamless and most satisfying manner. The first step involves watching and talking to users while they interact with the products targeted for innovation. But there are always questions as to how many users are enough for the research, do we need to consider the extreme users, who is our key persona amongst the various different users, which needs are we focusing on and which needs are not so important. These questions will have different answers for different product and user categories and thinking through them will lead to the design direction. It is very easy to be biased towards a particular user who makes a strong impression on you and forget about the rest of them which can result in a skewed design which works great for that particular user but really frustrates users with skill/interest levels above or below this user's level. It is interesting that even professional companies like IDEO only go to a handful (5-6) of users for key insights and create and iterate their designs based on limited feedback. I guess the key here is iteration...feedback from users definitely tells you what they NOT like, it gets more fuzzy when you want to put down what they DO like. So with iterations and different models, it becomes easier to get a set of values which are considered key in the design according to the user and build up on that. Ideally, there should be sufficient project time to do more than a couple of iterations to get close to user's expectations but then again if there's no time constraint, the process can practically continue forever.
These are just some random thoughts that I have with respect to research for UCD. I think that there are some general rules to it but you have to be ready for surprises and willing to take an entirely different direction if the users overwhelmingly seem to point towards it. Eliminating design alternatives based on user feeback (which might be often mixed) and still keeping the market success in mind makes decisions more difficult yet most critical towards the end of the UCD journey.
