The team casually assembles for the routine design critique. Stakeholders from the business and development teams join members of the design team as they prepare to review the latest concept designs and prototypes.
The designer kicks off the meeting with a greeting and a review of the problem, audience, goals, and other inputs that have influenced design decisions. Then it’s time to present the first design.
The team absorbs it quietly, along with her exposition. She pauses, anticipating the reaction.
What happens next is more than just a statement on her design. It’s a statement of the design values of the organization. Someone innocently says “I like it,” and the design team grimaces.
What’s so bad About “Like”?
When you tell someone you like something, it seems generous and charitable.
But, saying “I like your design” in a design review is, I believe, lazy at best, and at worst, lethal for a design practice. In my view it subjectively judges the superficial attributes of a design (often the aesthetics) and ignores the deeper quality — the purpose — which is its ability to perform or serve the customer.
This is why:
- It puts the emphasis on the evaluator and not the customer. It is self-indulgent and symptomatic of “inside-out” thinking; it is antithetical to customer-centric thinking.
- It is generic and lacks the specificity that allows for improvements.
- It is not a valid assessment of the design; it is subjective and is an opinion. (However, to the evaluator, it feels valid.)
- Positive subjective feedback provides a false sense of security; likewise, negative subjective feedback may focus attention on a feature unnecessarily and divert attention from other aspects of the design that need improvement.
- It sets the stage for a debate between those who “like it” and those who “don’t like it”. (This is the doorway to design purgatory.)
To be blunt, the opinions of designers and stakeholders are interesting, but irrelevant. It doesn’t matter what we think or feel about a design. “Like” is one simple word that can undermine a customer-centered, evidence-based practice.
Addressing Design Culture, Purpose and Performance
I used to manage the UX team at a multinational IT and networking company in Silicon Valley. There, our design team often joked that if you “liked” something, you had to put a dollar in a pot. We could have fed the team more than pizzas if we’d done this.
When someone slipped up and “liked” something, we promptly called each other out, because we wanted to change the language of design throughout the organization and we knew we had to start with ourselves. This shift in language was an important tactic in changing the design culture in the organization.
We put a moratorium on saying “I like…” to emphasize the importance to admire a design for the right reasons.
Since the goal of a design is to serve the customer, we wanted to admire designs based on how well they served customers.
For example, if we have two designs, we want to evaluate them based on performance criteria such as:
- Do customers adopt one faster than the other?
- Are they able to complete their task successfully?
- How long does it take?
- How many errors do they make?
- What features do they use first? How many features do they use?
- Is the value evident?
Most of these questions can be measured through objective methods of evaluation, such as observational studies or metrics. In addition, we can ask customers about their perceptions (such as value) and track their responses.
In short: a design must perform for a customer. Performance is measurable, and performance measurements are objective.
Therefore, it is better to use language in design reviews that addresses the performance qualities of the design and avoids subjective terms and assessments.
The opinion of the customer, as well as the behavior of the customer are valid measurements. This requires testing: it’s not as quick or cheap as an opinion from a team member or internal stakeholder, but it’s far more effective in improving designs, because it is specific and relevant to the customer.
How to Give Better Feedback
In the movie “Whiplash”, actor J.K. Simmons won an academy award for playing a demanding music instructor who famously said: “There are no two words in the English language more damaging than ‘good job’.”
While that may be overstating the principle, most literature on giving feedback – whether to children or employees – recommends you should be specific and avoid broad sweeping judgments.
These same principles also apply to feedback for designs.
- Be specific.
- Think about the intended outcome or behavioral change.
- Stay relevant.
Try to give feedback that considers the customer’s perspective and connect it to some customer insight based on evidence.
For example, instead of saying “I like the new button”, try “I think this button is more accessible to customers and will get more clicks, because we heard them use this word in conjunction to this action and we saw them examine this region of the page”.
The latter statement is still complimentary, but explains the rationale behind the compliment.
Quicker, Easier, More Seductive
Giving feedback like this takes more time and thought as well as knowledge. It also requires the evaluator to understand what the desired behavior is — in this case, to find a button and click it.
This is one of the reasons why it’s so tempting take the shortcut and simply state: “I like it”. It takes less cognitive effort, seems innocuous and feels natural to give our primal, subjective reaction to a design.
The next time you’re in a design review or critique, listen to yourself. Do you say “I like it”? Listen to others. Is the design being evaluated with likes and dislikes?
If you hear yourself or others slipping, take the opportunity to pause the discussion, reflect on the desired language for critiques, and redirect the discussion to evidence and insights gleaned from customer research.
It may seem small, but it’s an important step towards becoming a design practice that makes evidence-based decisions and methodically drives improvements through objective, empirical evaluations.
Try to change your behavior and the behavior of others by resisting the “liking” of designs.
By the time the customer says “I don’t like it”, it’s too late.