Critiques are one of the most important ingredients to great design, but far too often designers leave critiques secretly feeling scattered, frustrated, and disempowered. Even when everyone is well-meaning, there are many subtle ways a critique can go off the rails and hurt your design process.
Thankfully, with a bit of thoughtfulness and care, any company can craft a culture of productive, passionate, and candid feedback. Here’s how.
To deliver great design, teams need a culture of healthy feedback
I’ve worked on dozens of tech teams, each with a different culture of feedback. The worst feedback sessions felt like a trip to a target range — designers gently release their lovingly crafted work to the group, only to have it blown out of the sky by a thousand different complaints. (Why isn’t the button red? Can we make that bigger?) Other teams are overly polite and deferential. It’s easier, but silence won’t push designers from average solutions to great design.
I used to think that these differences in feedback style were baked so deeply into a company’s culture that they’d be impossible to change. But then I noticed something surprising: across all these different teams the feedback that was happening within design teams was always helpful. When designers gave feedback to each other, they were focused on the same goals, candid about where proposed solutions weren’t working, and creative about alternate approaches.
How did these designers create a microcosm of healthy feedback? It’s actually simple: they were all taught how to critique in school. Yep, it’s a totally learnable skill. And by following a few simple guidelines (and keeping each other honest to those guidelines) any team can build a healthy culture of design feedback.
1. Use formal critiques to jumpstart change
Any behavior change can be tricky. And getting a whole team to change how they give feedback might seem downright impossible. But a little structure can go a long way. Start by moving all informal design feedback into specific scheduled critique sessions. This allows you to invite the right people, articulate the goals, and lay out ground rules for feedback.
You should aim to have only a few people in the room so that there’s a single conversation. 5-6 people is about right. It’s also helpful to have a diverse set of experiences. Rather than include folks with big titles (CEO, CTO, etc.), think about who has the best data to inform the design. For example, including customer support or sales can help clarify customer goals. And including people familiar with the creative process can help when you’re feeling stuck.
Running the critique with a designated facilitator can help keep everything on track for the first few times. The facilitator’s job is not to give feedback, but to make sure everyone is giving the right kind of feedback. This might seem heavyweight at first, but as your team internalizes critique habits, you’ll need less structure. Soon every bit of feedback, even hallway conversations, will be more productive.
2. Start with critique guidelines
The most important idea for everyone to understand is the purpose of a critique: We are here to give the designer feedback so they can make the product better. We are not here to blame anyone for places where the product is failing, or to design better solutions on the spot. Critiques help designers improve their designs.
So start critiques by reiterating this purpose and reminding everyone how to give good feedback. Here’s a few guidelines that I’ve found helpful in critiques:
- Be candid — It doesn’t help anyone to stay silent during a critique, only to express your doubts privately later. Encourage feedback from everyone, whatever their title or role in the company.
- Be specific — Be as detailed as possible about what’s working and what’s not. If you say a whole design is not working, be prepared to back it up with a lot of specifics.
- Tie everything to goals — Critique not about whether you like or dislike a design. Good feedback is about how the design is meeting (or missing) the customer and business goals. Stay analytical. If you have an emotional reaction (this sucks!) dig for why you’re feeling that way.
- Affirm what’s working — Always call out what’s working well. Otherwise, you might lose a great idea in the next design iteration.
- Problems first, then solutions — If you think of a new design idea, great! So did everyone else though. So instead of arguing over solutions, start by taking a step back and first discussing problems with the current design. Then share your solution, smarty pants.
- Suggestions, not mandates — Any new ideas should be given as suggestions, not mandates. That’s because abstract ideas often seem better than concrete ones. Trust the designer to explore a new design direction and make their own call.
- Have fun! — When humans are stressed, we’re not very creative. But when we’re playing, it’s much easier to see the wide world of possibilities. So do anything you can to set the mood: play music as people are arriving, start with a funny video, throw around some NERF balls, have everyone stand, and pick a location that’s comfortable.
3. Set the stage
Even when everyone in a design critique knows how to give good feedback, things can still go off the rails. If a designer simply puts up a screen and says, “What do you think?”, the meeting is bound for chaos. That’s like jumping on a dogsled, saying “Mush!”, and then watching all the dogs take off at top speed in different directions because you forgot to put on the harness.
So after going over critique guidelines, but before presenting work, the designer must set the stage. Everyone is coming to the meeting with their own assumptions about customer needs, their own pet business goals, and differing familiarity with the project. Setting the stage is about creating a common world view and preparing everyone to see the design not in relation to their own experiences, but as a solution to a specific problem.
Here’s an example of how I typically set the stage:
- Review business goals — This quarter our team’s main goal is to improve conversions through the application process.
- Review customer goals — We’ve heard many times in customer interviews that the application process is too complicated and hard to complete in one sitting.
- Review constraints — The engineering team recommended a full rewrite anyway, so we have a lot of flexibility to make changes.
- Review schedule — We want to launch a new version in 2 months. So that means we should have a prototype nailed down for testing by next week.
- Check for agreement — Does this sound right? Anything I’m missing? We should hash it out now if anyone disagrees.
- Set expectations on level of fidelity — I’ve just spent a few hours sketching out an overall flow and some draft copy. So this is very rough.
- Direct the feedback — I want to know if you think this overall approach is right. Don’t worry about the layout of pages. I really need feedback on the flow and which features should be on which page.
4. Simulate the customer experience
Critiques are not pitch meetings. There are plenty of tricks for showing how clever a designer you are, or for getting a group to consensus. But this is a critique. We’re attempting to get inside the head of our customers, and see our product with fresh eyes. So for the first pass through the design, it’s helpful to simulate the customer experience as closely as possible.
- Pick a task — Instead of a single mockup, show a flow. Don’t assume your colleagues can imagine how the product might work. Pick a task and render every single screen. I know that’s a lot of work, but it’s worth it.
- Back it up — Start the flow a few steps earlier to get everyone into context. If you’re critiquing a mobile sign up, start with an app store. If you’re critiquing an email campaign, show a crowded inbox first.
- Use a screen — Paper is great, but your customers won’t be able to glance back at an earlier screen if they get confused. So for the first runthrough, it’s best to flip between screens.
- Fake it — Be scrappy about making the designs seem real. Throw your screens into Flinto or Keynote to make a rough prototype. Simple animations add a ton. And even if you don’t have hot-spots just pretend it’s working by moving your mouse. Trust me, no one will know.
- Don’t pitch hard — You won’t be there to pitch your brilliant design rationale for every customer. Your design needs to stand on its own. So start simply by explaining what the user is doing at each step. (You can break into a grid system discussion later.)
- Write down feedback — Ask everyone to silently write down feedback as you go. That way you can get through the whole task in real-time. You’ll also be gathering a wider range of opinions and avoiding groupthink.
5. Gather feedback and discuss
Now the easy part: Discuss! Go around the room and ask everyone for their feedback. Stay curious. And although we started by showing screens, it can be helpful now to move over to paper printouts so that people can see the whole flow, and so you can take notes directly on top of the designs. Don’t be precious about your printouts either — get in there and draw!
If the feedback gets a little harsh, it’s easy to take it personally. So if you start feeling bad, remember: you are not your design. Even great designers get off track. Re-focus your energy on listening and finding ways to improve the design in the next iteration.
You’ve got the feedback, now what?
Feedback is a gift. I love it. Whenever I’m too close to a problem, I rely on my coworkers to help me get perspective. But as a team, we can all be too close to the problem. So it’s important to get outside the building and show your prototypes to customers as soon as possible.
Critique is still the first way I get feedback on all my designs. And once I work with a team for a few weeks, this careful process of critique becomes easier: we can skip the guidelines, move faster through setting the stage, and even use the same structure for quick hallway discussions. All of those automatic habits slowly compile to be part of a team’s design culture. And once that strong culture is built, designers will be more productive, effective, and happy.
I know design critiques are a lot of work. They rob time away from other important tasks like writing code and replying to customer support tickets. But great design doesn’t come from putting a designer in the corner and hoping they come up with the right solution. The best designs spring from collaborations between product, engineering, customer support, sales, and others. Simply put: Design is everyone’s job. And if you’re looking for how to contribute to great design, start by getting your team to participate in design critiques.
Braden Kowitz is a product designer, prototyper, and storyteller. He advises startups on UX Design and Product Development. Before joining Google Ventures, Braden led design for several Google Products, including Gmail, Google Apps for Business, Google Spreadsheets, and Google Trends.
VB's research team is studying web-personalization... Chime in here, and we’ll share the results.