I also remembered some of the procedures of my previous job and I was actually very surprised to find out how much of it we had been doing wrong. Can't really change that now, since I've got a different job at a different company, but it was still really interesting for me to know that.
Here are my notes:
- Use more than one method; integrate them for your general process.
- You don’t get the same kind of knowledge from a market study graph than from actually observing users, but you can use both of the results.
- Innovating goes farther than just solving existing problems.
- Satisficing refers to acceptance of good-enough solutions.
- The knowledge “You’re not your user” creates empathy, but going out to the field makes you listen and understand what your users are going through.
- Through fieldwork you can detect unmet business goals.
- Doing fieldwork can accomplish many research goals at the same time, not only about the users but also about your organizational goals.
- Web analytics is never as descriptive as actual fieldwork
- Fieldwork should be used through the usual development cycle.
- At the beginning, to find out what you should do, what users need.
- In the middle, to make sure you’re doing the right thing
- Afterwards, to find innovation opportunities
- It’s important to have the knowledge of what you currently know and what you want to learn through your research.
- Plan and execute your research.
- Who to talk to? Recruiting users for the study, creating contrast to reveal key influencing factors.
- Consider the many existing types of users.
- Considering demographics is secondary
- What to do with them? Selecting methods to obtain the information you need through your study.
- Don’t directly ask people what they want, engage them in activities like participatory design or role-playing.
- Create a document to guide your session, even if you’re not going to follow it exactly.
- Do something with the data – analysis techniques
- Avoid jumping to conclusions
- There must be a clear distinction between your interpretation and what the user actually said. Both are important.
On a related subject, and for the spanish speaking audience, I recently published an article on the Software Guru magazine about defining persona profiles that could be helpful when executing the first part of your research (selecting your users). It's a printed article, but you can also view it digitally.
If you're interested in knowing more about the Web App Masters Tour this year, Luke Wroblewski has been publishing his notes too and they are much more detailed than mine. You should check them out.