You Should Invite Everyone to Your Design Critique

I recently attended a workshop on design leadership by Mia Blume where she described the difference between feedback and critique. She defines feedback as a quick, one-way reaction often grounded in opinion. Critique, she says, is critical dialogue grounded in objectives that helps us understand design decisions.

Unsurprisingly, most designers I've met want as little feedback as possible. It makes sense. The one-way nature of feedback doesn't give designers the time or space to provide context or articulate their decisions.

Meanwhile, people who don't have "Designer" in their job title are forced to make changes to design through feedback. Design isn't their craft, but they're heavily invested in the outcome. In fact, their jobs may depend on it.

When I plan a critique, I include Business Analysts, Product Owners, Subject Matter Experts, and Engineers along with Designers. I coach them on how to talk about design and give them guidelines to follow. Everyone learns from each other. Designers gain a better understanding of the business constraints. Engineers share technical ideas and limitations. All of this makes the critique more effective.

Here's what's on my screen during a remote design critique. Links to the the design are sent out so that people can ask questions and start dialogue even when there isn't a gap in the conversation.

These are the guidelines I give critique participants, building on what I learned from Mia Blume (take a workshop from her if you can!):

  • Let everyone participate. Every perspective on this call is important.
  • Ask questions if you’re unsure why something is the way it is.
  • Speak from objectives. Avoid phrases like “I don’t like that”.
  • Dig into the details. This is the time to talk through the nuances of word choice, button labels, error handling and more.

The whole team is responsible for what we create. By including them in the process, everyone has the opportunity to shape the design and the product is stronger as a result.