As user experience designers, we often come up against the misconception that user research is not important, that a design does not need to be fed by research, but that we can take the "genius" design route, assume that we know what we need to know, save a little time and money, and then evaluate our results via usability testing afterwards.
The problem is that this gives very hit and miss results and we quite often lose out on valuable insights and create a poorer experience by skipping the research step. Research can fuel innovation if it's done well.
Research is often thought of as a scientific process and this can lead to distractions such as focussing on methods and analysis techniques, at the expense of thinking about what the research is attempting to discover. We must focus on the problems we are trying to figure out, and the methods will come out of that.
Usability testing is very popular, it's fashionable in the web design/development community, but it only tells us the problems with the interface you put in front of the user, not whether you have understood the fundamental problem you are attempting to solve. Usability testing is evaluative.
User research gives you important new data on the problem itself, and the context you are investigating; it is generative.
User interviews (be they face-to-face, or via a questionnaire) are the most common form of user research, but it's easy to miss out on more innovative approaches, even to this familiar method. For example, Uscreates needed to engage migrant workers, so they set up a touring café campervan and set up outside the workplaces where they were most likely to find their target group.
Your approach can be adapted to fit the needs of the project. You might find you get exactly what you need from informal video taken from a mobile phone, or setting up a twitter/facebook account for your participants to post to. There are even new and inexpensive apps available to make collecting data even simpler, such as dscout, which allows you to set up questions and tasks for your participants and collect their responses along with a photo and date/time/location stamps.
There are plenty of methods out there, each with their pros and cons, so use what works best for the problem you are investigating. It doesn't need to be expensive, so there's no reason not to research.
Big reports are the tradition, but they are expensive, intimidating and not easily actionable for most people. So don't just collect results, analyse and interpret them; tailor the report to what will help the reader understand and take action.
For generative data, tell stories, help the reader empathise with the user, build personas to make the results come to life.
Work collaboratively with your client/researcher to find the solution to the problems you have found.
Lee wrapped up by presenting three simple take-aways from the talk: