Web App Masters: Uncovering Innovation with Fieldwork

by Luke Wroblewski May 24, 2011

At the Web App Masters Tour in Seattle, Steve Portigal detailed how insights from users can impact our designs and how to gather those insights. Here are my notes from his Design Fieldwork: Uncovering Innovation from the Outside In presentation:

  • Be a methods-polygamist. Choose, mash-up, or create a methodology based on the problem you are trying to solve. Integrate with other methods. Create a library of methods and artifacts that you can call on and modify as needed. Different methodologies tell you different things. It’s not an either or.
  • Innovation means going beyond pain points. If we focus on only the problems people know they have, we won’t really innovate. Diving deep is essential if we want to do more than solve known pain points.
  • Pain points might not really be that bad anyways. In many cases “good enough” solutions are all people need. Satisficing is the term that refers to our acceptance of good-enough solutions. What you see as a pain point, others might not. Looks for opportunities to delight and enable people in ways they can’t frame as a pain point.
  • “You are not your user” is a powerful way to empathize with people but it does not tell the whole story. Diving deep with fieldwork can reveal a lot more richness and common connections between customers and the teams building products.
  • Fieldwork can also uncover unmet organizational needs. Interviews reveal what people don’t know about an organization. Can also uncover actively used functions and how people believe they should be working. Fieldwork is great at discovering descriptions and perceptions.
  • When should you do field work? Early and often –use it throughout the development cycle. Can start at beginning to take a fresh look at what people do. During the refine and prototype phase can use fieldwork to look at existing ideas. Can explore new ideas when products are already in the market.
  • Huge issue is trying to figure out what problem you are trying to solve. What do we know and what do we want to know? Business goals tell you what the result will be. Tease out what you want to see happen. Research goals tell you what you want to learn. Don’t conflate research and business goals. The two are separate.

Who do you want to talk to?

  • What is the relationship of the people we want to talk with to the product/service/brand? This is different than just finding existing users. Look at a range of users to create contrast. Create contrast to reveal key influencing factors that you wouldn’t see otherwise. Examples: typical user non-user, extreme user, etc.
  • Think broadly about who the user is. There are probably more user types than you initially think. Think about the whole system: the chooser, the influencer, the user, and anyone who is impacted by those roles. This will surface who is affected by the product and who is being designed for.
  • Demographics are usually the last thing to consider in screening. But you can use them to create contrast in your panel.
  • Whoever the person you recruit, you want them to be engaged, have a point of view, and care about the thing. This will surface the issues you care about more. This is your research sample not your entire audience.
  • Screeners figure out if the person is a fit for your study but they also help to convince people to participate. Manage expectations right from the initial contact.
  • To recruit, you can work with an agency or do it yourself through networks, intercepts, craigslist, snowball recruiting (participants fond more participants)

What do you want to do with them?

  • Use a range of methods: interview, tasks, participation, demonstration, role-playing, blogging, homework, stimuli, and exercise.
  • Participatory design is not “solve the problem for me”. You need to uncover the need buried in the design request and decide what to do about the need. Not necessarily the proposed solution. Designers can work with this data to generate alternatives.
  • Consider the difference between testing and exploring. Don’t show your best guess at a solution but surface evocative images that illustrate the bigger issue.
  • Have a document that describes what you are going to do with your fieldwork. Create a pre-visualization of the session.
  • Workbooks are kind of survey-like but usually a lot more open-ended. They cover a range of activities designed to surface stimuli or get people thinking about topics.

Do something with the data

  • Between fieldwork and development there’s an aspect of synthesis and ideation. Too often, people jump to conclusions. Instead, you want to go through a process together.
  • Analysis: break large piece(s) into smaller ones in order to make sense. (interviews, transcripts, stories). Separate things and organize them. Synthesis: take the pieces you pulled out and put them into something new. The process gradually moves from analysis to synthesis.
  • Put time in to get value out of data. It does not reveal itself.
  • After fieldwork, collate reflections and quickly create a starter set of 5-10 thematic ideas.
  • Develop opportunities. They are not a report of interesting findings or a list of solutions. Opportunities are change we can envision based on what we heard and observed. We want to come up with lots of solutions through a generative process by asking “how can we...”
  • Report research in a way that starts to point to what can you do with it.
  • Small group ranking can be used to quickly work through ideas with a list of ranking factors that make selecting ideas more concrete.
  • You can tackle fieldwork quickly. Often in just a few days if your goal is building empathy, shorter times to analyze can help as well.