Stephen Brenish
Most bad art feedback is not malicious, it is untrained. “Make it pop.” “Can it feel more epic?” Notes like these are honest attempts to describe a reaction, but they give an artist nothing to act on. The revision comes back different rather than better, another round of notes follows, and a two-day task turns into two weeks.
As a producer, you usually are not the person giving creative direction on art; that belongs to the art director and leads. But you run the process that feedback flows through, you give plenty of it yourself, and you are the one who notices when revision cycles are eating the schedule. Good feedback is a production tool.
Describe what the art needs to do, not what it should look like
The most useful shift is from prescription to intent. “Make the sword bigger” is a solution. “Players can’t read the sword at combat distance, so they don’t know which enemy is the threat” is a problem. The second note lets the artist find the best fix, which might be scale, value contrast, silhouette or color, and teaches them something about the game in the process.
Anchor notes in the player’s experience:
- Readability: can the player tell what this is, and whether it matters, at gameplay distance and speed?
- Function: does it communicate what it does (interactive, dangerous, collectible, decorative)?
- Consistency: does it fit the style guide and sit comfortably next to the assets around it?
- Mood: does it support the feeling this moment of the game is meant to create?
Match the note to the asset’s stage
Feedback that is right for a final polish pass is wasted on a blockout, and vice versa. Detailed notes on an asset that might still be cut send hours of work in the wrong direction.
| Asset stage | What to give notes on | What to leave alone |
|---|---|---|
| Concept or rough | Idea, function, silhouette, fit with the brief | Detail, materials, finish |
| Blockout | Scale, proportion, readability in context | Surface detail |
| First pass | Shapes, materials, value and color relationships | Small polish issues |
| Second pass | Consistency with neighbouring assets, technical budgets | Revisiting decisions already signed off |
| Final polish | Specific, small, prioritized fixes | New direction |
If you find yourself giving polish notes on a blockout, stop and ask whether the asset is going to survive the next design decision.
How to run a review
Jobs in Game Development [Team Management] · Masahiro Sakurai on Creating Games on YouTube
Open by asking, not telling. Have the artist walk through what they were trying to solve. You find out whether the brief was understood, and you often discover a good reason behind something you were about to criticize. Sometimes the problem turns out to be the brief, not the execution.
Prioritize. Give the few notes that matter most, in order. A long list of equal-weight notes leaves the artist to guess what matters, and the important fix gets the same attention as the trivial one. Tag notes as blockers or suggestions if you are working asynchronously.
Separate blockers from preferences. “This breaks readability” and “I’d personally try a warmer palette” are not the same kind of note, so do not deliver them as if they were.
Close by confirming alignment. Before the session ends, both of you should be able to say in one sentence what the next revision is trying to achieve. If you can’t, the review is not finished. Follow up in writing with what was agreed.
Critique the work, not the person
The fastest way to damage an artist-producer relationship is a note that lands as a judgment of the person. “This looks amateur” in front of the team will get you safe, cautious work for weeks. The same concern, framed as “the lighting is fighting the silhouette here, so the character gets lost,” gets you a fix.
A few habits help:
- Give hard notes privately first, then discuss in the group.
- Talk about what the asset does, not about what the artist did.
- Say what is working as specifically as what is not, so the artist knows what to keep.
- When feedback comes over chat, add context. A screenshot with a red circle and no explanation leaves the artist guessing whether it is a blocker or a passing thought.
When you and the artist disagree
Check the brief first. If it was ambiguous, you each pictured something different and nobody caught it early. That is a production problem, and owning it changes the tone of the conversation.
Bring reference for the problem, not the solution. Not “make it look like this,” but “this is the readability problem I’m worried about; do any of these approaches help?”
Know whose call it is. If you are a producer rather than a creative or art director, your job is to make sure the disagreement is decided by the right person, on time, not to win it. Escalate to the art director with both views stated fairly.
Make written feedback unambiguous
For asynchronous reviews, use a tool that pins notes to a specific area or frame of the asset rather than describing locations in a message thread. Frame.io for video, Milanote or Miro for visual boards, and production tracking tools such as Autodesk Flow Production Tracking (formerly ShotGrid) for asset review all do this. Pinned notes remove a large share of the “which part do you mean?” back-and-forth.
A quick reference
| Feedback type | Use it when | Avoid it when |
|---|---|---|
| Intent note (“players need to read this as a threat”) | Almost always | Never |
| Prescriptive note (“make it 20% bigger”) | Late polish, with agreement on the goal | Early stages, or when you are guessing |
| Reference | Exploring a problem together | As a substitute for explaining the problem |
| Group review | Shared context, consistency across assets | Hard or personal notes |
| Written summary | After every review | As the only channel for difficult feedback |
For the wider team side of this, see managing creative feedback without killing morale and the creative director and producer relationship.
Photo: Yan Krukau via Pexels
