As the final post in this 4-part series, we will discuss Modification 4: Tactical user participation. Our previous posts (Part 1, Part 2 and Part 3) describe how a ‘modified scrum’ process can be an effective way to incorporate user-centered design techniques within an agile development process. To be consistent with our previous posts, we will compare best practices between traditional user-centered design and agile approaches, explain the modification and provide an example.
User-Centered Design Best Practice: Conduct thorough user research at the start of the project to better understand needs and refine requirements. Next, build prototypes early and do usability testing. Finally, perform another round of usability testing after production code is ready. Capture as much information from users as you can during those limited number of interactions with them.
Agile Best Practice: Bring users into weekly meetings to help flesh out and prioritize the user stories.
The traditional user-centered design process provides a limited number of interactions with users. And since it is often not clear what you want to learn from the users, a lot of time is required to plan the research sessions and analyze the results. On the flip side, during an agile project there are multiple sprints, enabling many opportunities to interact with users. But in reality, for complex consumer or business applications it’s impractical to have users participate in so many discussions—much of the technical information will likely be way over their heads.
Modification 4: Tactical user participation
User research is performed during every iteration, bringing users into the design process in a tactical manner.
Why this modification?
By tactically bringing users into the design process during every sprint, designers have the opportunity to learn valuable insights from users every 3-4 weeks (or the length of the iteration cycle). Designers can focus on small, bite-sized segments and test only current activities at that point in time, rather than creating big monolithic research plans (with the fear of not being able to reach users.) If new interactions are created after users are involved in the current sprint, designers can be confident they will be able to test and learn more in the next sprint.
A hotel search results page needs to be redesigned for a travel Web site. Designers have several items they would like to discuss with users. The next user touch point is scheduled a week from now. Users are actively recruited even if the subject is currently unknown. All the designers suggest what items they would like user input on, and then prioritize and choose one. In this example, the designers chose to gather what information is important to consumers. For example, is the price, star rating, address or the fact that the hotel has a pool important?
In the next iteration, the same pattern occurs. Designers enjoy a great level of comfort knowing there is regularity in the opportunity for feedback. And they can effectively prioritize the most important items at a given point in time, knowing future feedback sessions will occur in subsequent sprints.
As a conclusion to this series, you can successfully implement great user experience design within an agile development framework. By effectively marrying the two disciplines, both designers and developers make compromises that can lead to significant business results.