User Research, Task Analysis, & Sketching

Due: Monday, March, 25th at 11:59AM

Assignment Overview

Your assignment is to use the contextual inquiry method to learn more about the practices of your target audience. Based on insights from the contextual inquiry, you will develop a set of tasks and sketch some initial interface designs.

At a high level, this assignment involves the following steps:
  1. Find justifiably relevant users to perform contextual inquiry with
  2. Analyze data from contextual inquiry to formulate tasks that users' want to accomplish
  3. Categorize these tasks into easy, medium, and hard tasks from the user's point of view
  4. Choose three of these tasks (one from each level difficulty level) to write about
  5. Sketch out interface ideas to support these tasks. remember, the task and the implementation of the interface are two different things. you could come up with lots of different interface manifestations to support a single task.

What to Do

1. Interview and observe at least four target clients/users using the contextual inquiry method. Follow the guidance provided in the IDEO Method Cards, "An Ethnographic Approach to Design" by Blomberg and Burrell, and Contextual Design by Beyer and Holtzblatt. Each team member should attend at least one contextual inquiry session.

2. Develop three tasks informed by the results of your contextual inquiry and using the task analysis approach. Select tasks that participants will perform with your application and that capture the important aspects of the problem you are solving. Include one easy, one moderate, and one difficult task. These tasks may include existing ones you observed as well as new tasks you’re your proposed solution will enable.

Remember that tasks say what is accomplished, while leaving open how to accomplish it—that is, the tasks should be about the user’s real world goals and should not focus on any specific design or interface. Keep in mind that even though you’re writing up three tasks for this assignment, you can and should consider revising your tasks in the future as your understanding of the problem matures.

3. Brainstorm and sketch three very different initial designs for your interface to accomplish each task. Do not illustrate the entire interface, but instead sketch key screens needed to illustrate the functionality (perhaps 3-5 screens for each of your three design ideas). These should be rough sketches, not polished artwork. They should include illustrations of the relations between screens or actions with the interface (e.g., conceptual illustrations of the design, arrows showing relationships between screens). Note: it's best to come up with the three different design aesthetics/approaches before creating the interfaces to accomplish each task. The key is to force yourself to think creatively about how these interfaces could be built. Do not allow yourselves to get functionally fixated on one design.

What to Hand In

You will submit a proposal of no more than 5 pages of text, approximately 2500 words. Images are strongly encouraged, do not count against the page limit, and are thus effectively free (the limit applies to the approximate amount of text you would have if all images were removed). Your reference list is also free.

Your submission must be in PDF format and linked to on your team wiki.

Your report should follow this outline, and will be graded using the guidelines discussed at the end of this document. The provided page allocations are rough estimates, to help convey how to divide up the space.

  1. Project title (something short and catchy to capture the key idea)
  2. Each team member’s name and role
  3. Problem and solution overview (short, 1 paragraph)
  4. Related Work (short, 1-2 paragraphs)
  5. Contextual inquiry participants (.75 page)
  6. Contextual inquiry results (1.25 pages)
  7. Analysis of existing and new tasks, including three supported tasks (2 pages)
  8. References (not part of page limit)
  9. Additional sketches of design (not part of page limit)

Writing Guidelines and Grading

Overall Writing Quality (10 pts)
Make sure your writing is easy to read: ensure it is clear and concise, use section headings, make liberal use of whitespace, include images in the body of the write-up with appropriate figure numbers and captions, refer to the figures in the body of your text, and check for grammatical errors.

Problem and Solution Overview (10 pts)

This overview should be a concise statement of the problem you are tackling and a brief synopsis of your proposed solution. Both the problem and the solution will likely have evolved since your initial team project proposal.

Related Work (10 pts)

Provide a summary of related commercial products or research papers. When referring to past work, it’s important to include a description of the past work, why it’s relevant and significant to the proposal at hand, and what opportunities or areas for improvement exist. This summary can be based on the team project proposal, but you can (and should) improve the related work based on the project’s evolution (and feedback from us).

Contextual Inquiry Participants (15 pts)

Describe the rationale behind your choice of target contextual inquiry participants. For each of the three (or more) participants, give some details of their background, the environment where you observed their practice, and your role as the “apprentice.” For this report, don’t include personally identifiable information on participants, such as their full name. Instead, you could use first names only, pseudonyms or initials.

Contextual Inquiry Results (25 pts)

Identify high level tasks and themes the participants shared in common in their practices. Note anything unique about each interview and comment on the rationale behind these events.

Existing and New Tasks (20 pts )

Describe and analyze three tasks, including existing tasks participants already do and/or tasks that will be enabled by your design. These should be real world tasks that have details (e.g., programming a DVR to record the Simpsons on Sundays).

Sketches (10 pts)

Sketch three different ideas that seem plausible, but that each take a very different approach to this design (for each of the tasks: easy, medium, and hard). The sketches should be rough, done on paper and then scanned (do not create or recreate them in a drawing package). You will be graded on the quality and diversity of these ideas, as well as the execution (sketches are rough, transitions are clear, there is enough to get each idea across). The sketches should convey your visual thinking about how you might solve this problem.


Below are some questions and answers from the first time I taught this class. It is my hope that this can help you avoid the same confusions/mistakes that some teams experienced last year.

What are "tasks" and how do they relate to our interactive system?

Some context: one team was working on a project trying to increase parental involvement in student education. They said:

For the three different interface design part of the project we are taking the following three approaches -
  1. Parent follows a wizard in the app to create his account, registers his children with the system, then goes to the dashboard page
  2. Parent already has an account, logs in, registers his children with the system, then goes to the dashboard page
  3. Parent already has an account, children are already registered with the system, logs in and goes to dashboard page

Are these approaches sufficiently different according to the assignment requirement?

My response:
I think these are too similar--in fact, they largely describe the same overall "task" which is logging in and viewing the dashboard page. Instead, it's better to think about what sort of tasks your system is trying to support--that is, what kind of tasks can users perform once logged in. Use your contextual inquiry to inform these tasks. For example, perhaps you learned from one of your interviews that parents wanted a way to view upcoming exams--how would you sketch an interface to support this task? Or, perhaps you learned that parents want to learn more about their child's absences, or how their child compares to other children in the class, or how their child is improving in a class, etc. Again, drawing upon data that you gathered in your contextual inquiry, you come up with tasks and then some design sketches to support these tasks.

Remember tasks essentially describe things that people do (or want to do) independent of any interface--think of them sort of like goals. Then, once you identify tasks, your job is to brainstorm and create sketches to support those tasks. There are many ways in which an interface could be designed to support any one task. For example, if we are making a music player application and we determine that our users want to be able to make playlists, we could generate lots of different interface sketches to support that same task.

What does "identify high level tasks and themes the participants shared in common" mean?

The whole goal of contextual inquiry is to investigate potential users in their own context and better understand their goals, tasks, and current practices. You then analyze and synthesize data from these investigations and identify commonalities across your users. From this, you determine that your interface must support X and Y tasks.

For example, let's say we are involved with a company building a new kind of billing system for dentists. So, we decide to go out and interview and observe dental receptionists to see how they currently handle the practice of billing (this is our contextual inquiry). For most of our clients, we observe that the receptionists use computers to generate the billing records (as expected) but then we notice that they actually print out each and every bill onto paper so that they can make handwritten notes next to different fields on the bill. This practice is inefficient and costs money (plus, it's hard to store all these physical paper records). Consequently, we decide that our digital system needs to support an easy way to annotate any part of a digital bill. This also leads us to think that an iPad or slate device might be a good alternative platform for the receptionists to use in their work.

Remember that contextual inquiry is just one part of the formative inquiry process--other parts may include researching competitive products, interviewing experts, and performing many of the techniques outlined by the IDEO cards.

What are "existing" vs. "new" tasks?

You only need to describe three tasks in detail--those could be existing tasks, a mixture of existing tasks and new tasks, or three new tasks enabled by your application. The key here is the rationale that you provide for these tasks that is, hopefully, tied back to your contextual inquiry.