A model for UX design reviews

Design reviews are so important for our work as user experience designers, but they too often fail us. Here is a model for design reviews that overcomes the problems of ego, emotion, and communication that so often get in the way of helpful feedback.

Alaina Kraus, Caitlin Potts, and I presented this process for the Jan 27, 2011 IxDA Lansing meeting.

Roles in the design review process

  1. Designer
  2. Facilitator (may also be a reviewer)
  3. Reviewers (you’ll need at least 2, but 6 may be too many)

Step 1: Designer explains the project and design concerns

At this step the designer shows the design (on paper, on screen, a prototype, etc.) and explains to the reviewers the context of this design. The designer should also point out specific areas that she wants feedback on.

Step 2: Reviewers discuss the design

This is the part that’s going to go against your habits.

The designer steps back and withdraws from the discussion. She should use body language to exclude herself to make it harder for the reviewers to address her. She should instead focus on her listening and note-taking.

Meanwhile, the reviewers are discussing the design amongst themselves. Instead of referring to the designer, they should refer to the design. Instead of talking to the designer, they should talk to each other.

Of course, the reviewers should point out good aspects of the design as well as discuss areas that they’d like to see improved.

By intentionally excluding the designer from this conversation, the dialogue can cover more ground and the feedback can be more honest. Otherwise, the dialogue has a good chance of focusing on a single point or so as the designer begins to explain or defend her decisions. That will derail good feedback, and everyone loses out on good information. People’s feelings are also at risk.

Step 3: The designer rejoins the conversation.

Finally, once the facilitator has determined the design has been discussed enough, he will invite the designer back into the discussion.

The designer can now summarize the notes she has taken in order to give the reviewers the opportunity to catch any misinterpretations. She can also ask follow up questions to clarify feedback she may not have fully understood.

The designer will probably want to defend the work or explain it, but really doesn’t have to. The goal is feedback for her, and at this point, she has it.

Why this method works

This method works because it makes it okay for the designer to simply listen in without feeling a need to defend her work. Likewise, it frees up the reviewers to not have to worry about hurting the designer’s feelings or fear the reaction of feedback taken poorly. While the method may feel a little awkward at first, after a few times it becomes easier.

What kinds of feedback make sense?

The reviewers should provide feedback that matches the fidelity of the design. This is to say, if it’s a rough sketch or task flow diagram, talking about pixel-perfect alignment of the layout is inappropriate. This should be obvious to most of us.

Also, instead of simply presenting your opinion about a design, discuss it in terms of usability heuristics (Nielsen’s list of 10, Tognazzini’s 1st principles document), accessibility concerns, visual design principles (e.g., proximity, alignment, repetition, contrast), and your observations from usability tests.

Where did this method come from?

In the mid-90s I worked at the Michigan State University Writing Center, and we used a similar process that we called the “fishbowl” to teach people to do peer-review writing workshops. In many ways, writing and designing are similar. When I started doing more design work, I recalled this process and adapted it to design reviews. It seems to work great. I credit learning this method from Dr. Sharon Thomas and Dr. Laura Julier of Michigan State University.

Update Feb 26, 2011: Wordcast live on Design Critique

Tim Keirnan over at the Design Critique podcast posted an interview we did on this process. Check it out! And thanks, Tim, for inviting me.

Update July 14, 2011: Michigan UPA workshop

This evening Alaina, Caitlin and I ran this workshop for a Michigan UPA event in Lansing. We had fun and it sounded like the attendees enjoyed themselves while learning this model. Thanks again Second Gear Coworking for letting us use your excellent venue!

6 thoughts on “A model for UX design reviews

  1. Last night’s design review workshop was excellent, Davin! There were two things that struck me as particularly important to the process. The first is the role of the facilitator. It seem like the ability of the facilitator to set boundaries, encourage constructive feedback and establish expectations for the group is critical in the success or failure of the review. I actually think the facilitator could be the most important role, if the end goal is to achieve truly valuable and honest feedback. The second part that struck me as particularly important was the third step where the designer confirms the feedback given. Beyond clarifying questions about particular comments, I see this as an opportunity for the designer to effectively say to the review group, “This is what I heard you say.” Even if the designer has no questions for the reviewers, I think it’s important for her to recap what she heard. This echoing back can be extremely important to the process since we can all interpret feedback in different ways, especially when it’s about something in which we are personally invested.

  2. Great article. I’ve been thinking a lot about design review and critique recently. It seems like what you talking about could be extended to portfolio reviews as well. As an aside, I’m helping to coordinate the upcoming Midwest UX conference in Columbus, OH on April 9-10. Any interest in submitting a proposal? http://www.midwestuxconference.com/

  3. *Incredible* synergy — just yesterday I pulled literature from a seminar I’d attended in 2009 on collaborative leadership (from the Social Interaction Institute for Social Change in Boston) on their version of this methodology, which they refer to as the Consultancy Model. Like you Davin, and your IxDA Lansing cohort, I was thinking of it in terms of design critique and feedback! I’ve used this method very successfully in non-IxD contexts (community organizing and project design). Excited to see the idea sprouted up sort of simultaneously in Michigan and San Francisco :)

  4. Chris, Thanks! It was a fun night. The session helped me appreciate the role of facilitator even more as well. Good points.

    Erik, the Midwest UX Conference looks to be shaping up nicely! The speakers look excellent. This design review exercise works really well as a workshop, so sure I’ll think about a proposal. Thanks for letting me know about it.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>