Do you want critique, or a hug? How to gain valuable criticism on your design

One of the most fundamental parts of the process of design is the critique, a formal opportunity for the designer to receive feedback from a group of people. There's a lack of good literature on critique, although there are a few notable exceptions, and so for most, critique remains a mysterious tool. Those of us who went to art or design school learned how to do it, but likely never learned explicitly; instead, it was much more of an experiential process. I remember showing up at my first critique at CMU and being completely mortified by the thought of putting our assignment (a self-portrait) on the wall in front of everyone else, and then talking about it.

I wrote about critique in a broad manner for interactions a year ago, but I've been thinking about more specific ways to introduce students to the idea of being critiqued. Here are some thoughts about how to receive criticism; I'll assume that the critique session is actually well organized and not just people sitting around talking, although that might be a poor assumption.

Be quiet.

When you are receiving a critique, it's extremely tempting to rationalize your design decisions—to explain why you did the things you did. This will always come across as defensive, because it is: your rationalization is actually a way of showing that you've thought through the trajectory of the conversation and already considered (and judged) the end state. The defensive quality of a rationalization changes the conversation from a way to produce new knowledge to a verbal debate. But you've already chosen the medium to make your argument, and it's your actual design work. By moving from the artifact to words, you game the system: your users won't have access to your words when they receive your argument in the form of the final design. All they have is the thing you've made, and so it needs to offer the complete argument on its own.

Additionally, when you rationalize and describe design decisions prior to critique, you steer the conversation. For example, if you begin by explaining your color choices, you've done two things: called attention to a particular design element, at the expense of the whole (and primed the group to be thinking mostly of color and aesthetics), and set up a boundary around your design choice. Some people refuse to cross these boundaries once they've been publicly established, because you've implicitly claimed ownership over a design detail: you've signaled to the group "I care about this, and if you poke at it, you'll hurt my feelings." Ironically, you may have called attention to it because it's the element you are most concerned about!

Write it all down.

Some of the best parts of a critique come from the small, nuanced details of conversation, and the ideas sparked by the conversation. A participant might say something like "When the user clicks here, instead of going to that other page, it seems like we could do a mini-modal, eliminate a step, and provide a way for them to maintain an understanding of context." There's at least three points that are important here (a stylistic decision of using a small, in-line modal; an implicit recommendation that the flow is too long; and an observation that context is important to make decisions). It's unlikely that you'll remember all of that when the critique is over, and if it's a good critique, it's unlikely you'll remember anything, because you'll be actively considering so many new and different ideas. It's critical that you write it all down.

When my own work is being critiqued, I number each individual item, component, or artifact of the design with a unique identifier. Then, as a person is speaking, I try to type exactly what they say into a document, and link their comment to the design prompt with the unique identifier. I also try to log who said what, so I can follow-up later if I need to. In a few instances, I've found written feedback to be politically useful, too—when teams wonder where seemingly irrelevant design decisions came from, it's effective to be able to point back to the origin of the idea as coming from within the team.

Extract more details.

Talking about interaction design is hard, because it's multi-layered, requires an understanding of the system, and is highly contextual. It almost always requires a dialogue to really understand any criticism that's offered. The dialogue, however, is a chance for you to ignore the first suggestion (Be quiet), and so while it's necessary to ask for clarification, it's important to do it in a neutral and open-ended fashion.

Consider two different approaches:

The second choice seems light and almost therapeutic; it's entirely non-confrontational, and acts in a way similar to the "five whys" of identifying root causes. While the first approach—"Why do you want to eliminate a step?"—is objective, it won't be perceived that way. Most people will hear you say "I don't want to eliminate a step. Why do you want to eliminate a step?" and you'll be herding them into a defensive corner and changing the tenor of the conversation.

Reserve time for conflict, and realize that you don't have to agree.

Just because someone said something, and you wrote it down, doesn't mean that you have to act on it. A critique is not a mandate. But be warned that there's something strange that happens in meetings: people leave with very different views about what happened. If you are quiet during the critique, scribbling notes, people will leave the session feeling validated—that you heard them—and expect that their comments will be illustrated in the next round of revisions you do. And they'll be personally frustrated when they don't see the changes they described, because they'll feel like you ignored them and they wasted their time. At the end of the critique, it's critical that you set expectations about what you intend to change, and why you intend to change it. This is hard, though, because you might not know at that point, and your comments will likely open the door for further discussion (which takes time). It can be effective to end with a simple phrase like "I heard all of what you said, and wrote it all down. You've given me a lot to think about. I don't agree with everything that was said, and so you may not see your comments visualized in the next iteration. If you feel strongly about what you said today, let's try to talk about it in a one-on-one setting." In large and politically volatile groups, I would recommend actually emailing both your notes and this disclaimer to everyone that was in attendance, and be prepared to explain—in a presentation or work session, but not in a future critique—why you made choices to ignore design suggestions.

Don't ask for critique if you only want validation. If you want a hug, just ask.

A "bad critique" is one of the most valuable things a designer can receive, because it short-circuits the expert blindspot and helps you see things in new and unique ways, and it does it quickly. But sometimes in the design process, you don't actually want feedback at all: you want affirmation, and you want someone to celebrate your work so you feel good. Learning to understand the difference is critical, because if you ask for critique, people will give you critique. But if you ask people to tell you the three best parts of your design, they'll probably do it. As Adam Connor offered in his IA Summit talk, "Don't ask for critique if you only want validation. If you want a hug, just ask."

Originally posted on Mon, 30 Apr 2012