Your design team doesn’t need more feedback. It needs better decision rules
Why strong creative direction is not about giving more comments - it is about teaching teams how to judge work before the review begins
Design teams rarely suffer from a complete lack of feedback. In most organizations, feedback is everywhere. Comments in Figma. Slack threads. Meetings. Screenshots with arrows. Stakeholder reactions. Last-minute notes before launch. The problem is that volume does not make feedback useful. A designer can receive twenty comments and still leave a review without understanding the most important thing: what makes one direction stronger than another?
I have seen this become especially visible as teams move faster. When production accelerates, review cycles multiply. More work is created, more people can comment on it, and more opinions enter the process. Without a shared decision system, creative direction can quietly turn into approval dependency: designers produce, leaders react, designers revise, leaders react again. The work moves, but the team does not necessarily become better at making decisions.
As a Head of Design, that is exactly the pattern I try to avoid. I do not want a team that can only produce strong work when I am present in every review. I want designers to understand the logic behind the decisions well enough to make more of them independently.
The goal of creative direction is not to become the person with the most comments. It is to make good judgment more transferable.
“Make it more premium” is a reaction, not a decision rule
Some of the most common phrases in design reviews sound useful because everyone understands the emotion behind them.
“Make it more premium.” “This one feels stronger.” “It needs more energy.” “I prefer version B.” “Can we make it pop?”
None of these statements are inherently wrong. They describe a reaction. But a reaction and useful creative direction are not the same thing. If “premium” means less visual noise, stronger typography, more controlled spacing, restrained use of effects, and a clearer hierarchy, the designer can work with that. If “premium” simply means “I will recognize it when I see it,” the designer has to guess.
That difference matters because repeated guessing creates dependency. The designer learns which version the reviewer prefers, but not necessarily what principle produced the preference.
Good critique should analyze whether the work meets its objectives, rather than simply judge whether someone likes it. Nielsen Norman Group defines design critique in exactly this way: as analyzing a design and giving feedback on whether it meets its goals. Design Critiques: Encourage a Positive Culture to Improve Products
The strongest feedback converts reaction into reasoning.
Instead of:
This doesn’t feel premium enough.
Try:
The visual hierarchy is competing with the decorative treatment. I would simplify the secondary elements so the product and message feel more intentional.
Instead of:
I like the other one better.
Try:
The other direction communicates the campaign proposition faster and has a clearer path for adaptation across formats.
The difference is small in language and enormous in what the designer learns.
A review needs a purpose before it needs opinions
One reason design reviews become chaotic is that different people enter the meeting believing they are reviewing different things.
One person is evaluating the concept. Another is checking brand consistency. Someone is discussing copy. Someone else notices a production issue. A stakeholder is effectively deciding whether the campaign should exist at all.
All of these conversations may be legitimate. They simply should not happen accidentally at the same level.
Before a review begins, I find it useful to answer four questions:
What stage is the work in? Are we exploring a direction, selecting a concept, refining execution, or checking production readiness?
What decision needs to be made? Are we choosing between routes, solving one specific problem, or approving final execution?
What feedback is useful right now? Conceptual, visual, technical, brand, production, or something else?
Who owns the final decision? Is the presenter collecting perspectives, or is this an approval meeting?
This sounds basic, but it changes the quality of the conversation immediately.
Figma’s own design team came to a similar conclusion when rebuilding its critique process. One of its core practices is to align on the goal of the critique and explicitly tell participants what feedback the presenter is — and is not — looking for. How we do design critiques at Figma
Without that framing, feedback expands to fill the meeting.
Separate taste from criteria
Taste matters in design. Pretending otherwise does not make teams more objective.
Experience changes what people notice. Senior designers often recognize imbalance, generic visual language, weak hierarchy, or an inappropriate emotional tone before they can fully articulate why.
The problem begins when taste becomes the final explanation.
A healthy design culture does not try to eliminate subjective judgment. It gives that judgment a structure.
When I review work, I try to move the conversation from:
What do I like?
toward:
What are we trying to achieve, and which direction does it best?
That changes the leader’s role. Instead of acting as the final aesthetic authority, the leader becomes responsible for making the criteria visible. If a campaign should feel restrained, explain what restraint means in that brand. If a direction needs stronger product presence, define what is currently competing with it. If experimentation is encouraged, define what can change without making the brand unrecognizable. If consistency matters, identify which elements are actually non-negotiable.
The better these rules are defined, the less often designers need to ask, “Is this okay?” That is a much more useful measure of design maturity than how quickly a team can execute revisions.
Give feedback at the right level
Another common problem is giving execution feedback when the actual issue is conceptual.
A team presents three campaign directions.
The real problem is that none of them communicate the proposition clearly.
The review begins discussing shadows, type sizes, image crops, and button placement.
The work improves technically while the fundamental problem survives.
Get Kateryna Tskhovrebashvili’s stories in your inbox
Join Medium for free to get updates from this writer.
I think of feedback as operating at different levels:
Problem level: Are we solving the correct problem?
Concept level: Is the idea strong enough to solve it?
System level: Does the direction belong to the brand and work across the ecosystem?
Execution level: Is hierarchy, composition, typography, imagery, and craft resolved?
Production level: Can this actually ship correctly across formats, markets, and technical constraints?
A lower-level correction should not be used to hide a higher-level weakness.
There is little value in perfecting spacing in a direction that should not exist.
This is also why the timing of critique matters. Late-stage reviews can surface problems when the cost of changing direction is already high. Figma’s engineering team has described the same issue in technical reviews: feedback becomes much more disruptive when it arrives after a direction is already built out.
Creative teams have the same problem.
Review the idea while changing the idea is still cheap.
The review should make the next review easier
A strong critique does more than improve one asset.
It creates reusable knowledge.
Imagine that a designer repeatedly receives this feedback:
- the product becomes secondary to the visual effect;
- the composition loses clarity in smaller formats;
- campaign concepts rely too heavily on generic luxury cues;
- localization breaks the hierarchy;
- experimentation changes too many brand signals at once.
If each comment remains attached only to the specific file, the team will rediscover the same lessons repeatedly.
If those patterns are documented, they become part of the design system.
This is why I believe design teams should document not only what they approve, but also what they repeatedly reject.
Not every rejected concept needs a case study. A simple screenshot and one sentence explaining the failure can be enough.
Over time, the team builds a shared memory:
This is what “too far” looks like.
This is where the brand stops feeling like us.
This is why this type of composition fails at CRM size.
This visual treatment looks impressive but consistently weakens communication.
The purpose is not to restrict creativity.
It is to prevent teams from paying for the same lesson twice.
Feedback does not end when the meeting ends
One of the easiest ways to damage feedback culture is to ask people for critique and then never explain what happened to it.
A designer receives ten comments. Some are implemented. Some are ignored. Some contradict each other. Then everyone moves on.
The next time people are invited to critique, they have no idea whether their input mattered.
A recent Nielsen Norman Group article makes an important point: critique needs a closed feedback loop. Teams benefit from communicating what changed after the review, what did not change, and why. Closing the Loop: What to Do After a Design Critique Ends
This does not require another meeting.
Sometimes a short message is enough:
We kept direction B because it communicated the proposition most clearly. We used the feedback about hierarchy and reduced the secondary visual elements. We did not change the photography treatment because it is part of the campaign system we are testing across the full launch.
That closes the loop. More importantly, it shows how decisions are made.
AI can give feedback. That makes human review systems more important.
There is another reason teams need clearer decision rules now.
AI is beginning to participate in critique itself.
Figma’s current AI design-review features can review selected work for areas such as visual hierarchy, accessibility, usability, and brand consistency, summarize comment threads, and provide feedback directly in the design canvas. This can be extremely useful. It also creates a new question:
If feedback becomes abundant and instantaneous, what determines which feedback matters? The same principle applies. More feedback is not necessarily better feedback. A tool can identify inconsistencies. It can surface accessibility problems. It can suggest hierarchy changes.
But the team still needs to define the intent against which those suggestions are evaluated. AI makes critique easier to access. Leadership makes critique coherent.
The goal is not better approval. It is better judgment.
There is a version of design leadership where the leader becomes the quality-control layer for everything. Every important asset waits for their reaction. Every difficult decision moves upward. Every designer eventually learns what the leader likes. This can produce good work. It does not necessarily produce a strong team. The more scalable version of leadership is harder. It requires explaining why. It requires making standards explicit. It requires letting designers make decisions and occasionally make the wrong ones. It requires building a review culture where disagreement is useful rather than threatening.
And it requires accepting that the ultimate proof of good creative direction may be that fewer decisions need to reach you.
A mature design team should not become better at predicting the Head of Design’s preferences. It should become better at evaluating the work itself.
Good creative direction improves the work in front of you. Great creative direction improves the decisions that come after it.
