A good design critique is one of the most valuable activities a design team can do. A bad one is a waste of an hour and a minor blow to morale. The difference is almost entirely structural.
Set the frame before you show the work
The person presenting should spend two to three minutes setting context before showing a single frame. What problem were you solving? Who is the user? What constraints were you working within? What specific feedback are you looking for?
Without this frame, reviewers default to evaluating personal preference rather than fitness for purpose. The frame converts 'I don't like it' into 'does this solve the stated problem?'
Separate observations from suggestions
The most productive critiques distinguish between what reviewers observe (this section feels heavy; the call to action is easy to miss) and what they suggest (make it lighter; move the button). Observations are data. Suggestions are hypotheses.
Train your team to lead with observations. The designer can then evaluate the observation against their own understanding and decide whether a suggestion follows. This preserves the designer's decision-making authority while still benefiting from fresh eyes.
Time-box and rotate facilitation
A 45-minute critique with a hard stop is more productive than an open-ended session. Use the first ten minutes for context, twenty minutes for structured feedback, and fifteen for discussion and prioritisation.
Rotate the facilitator role. When designers facilitate their own critiques they tend to defend rather than absorb. An external facilitator — even a junior team member — keeps the energy on the work rather than the person.
Document and follow up
A critique that produces no record is a critique that will be repeated. Capture the key observations, the questions raised, and the decisions made. A three-bullet summary shared in Slack after the session takes five minutes and saves hours of repeated alignment.
Structured Design Critique: Key Practices
- Set the frame before showing work
- Spend 2–3 minutes on context: problem, user, constraints, and the specific feedback you want.
- This shifts feedback from personal preference (“I don’t like that colour”) to purpose (“does this solve the stated problem within these constraints?”).
- Separate observations from suggestions
- Observations = data (“this section feels heavy”).
- Suggestions = hypotheses (“maybe make it lighter”).
- Coach the pattern: “I notice…” (observation) and optionally “…which makes me wonder if…” (hypothesis).
- The designer keeps decision-making authority; critique provides inputs, not directives.
- Address the limits of peer-only critique
- Peer critiques among designers tend to over-focus on aesthetics because of shared craft vocabulary.
- Include PMs and engineers with a clear brief on what to look for (e.g., edge cases, data states, outcomes).
- Questions like “how does this behave with empty data?” often add more value than purely visual tweaks.
- Time-box and rotate facilitation
- Aim for ~45 minutes with a hard stop:
- 10 minutes: context
- 20 minutes: structured feedback
- 15 minutes: discussion and prioritisation
- Protect the context phase; without it, feedback devolves into opinion.
- Rotate facilitators so the presenting designer can listen instead of defend.
- The facilitator guards the frame, cuts rabbit holes, and ensures broad participation.
- Match critique format to fidelity
- Early-stage, rough work → smaller group, generative, exploratory.
- Late-stage, near-final work → broader group, focused, criteria-based feedback and sign-off.
- Avoid running full formal critiques on sketches or loose brainstorms on almost-final designs.
- Document and follow up
- Capture key observations, questions, and decisions; even a 3-bullet Slack summary is enough.
- Documentation prevents repeating the same critique and shows designers their input was heard.
- Turning discussion into written decisions is the discipline that makes critiques impactful.
Structured Design Critique: Summary
1. Set the frame before showing work
- Spend 2–3 minutes on context: problem, user, constraints, and the specific feedback you want.
- This shifts feedback from personal taste (“I don’t like it”) to purpose (“does this solve the stated problem within these constraints?”).
2. Separate observations from suggestions
- Observations = data (“this section feels heavy”, “the CTA is easy to miss”).
- Suggestions = hypotheses (“make it lighter”, “move the button”).
- Coach the pattern: “I notice…” (observation) and optionally “…which makes me wonder if…” (hypothesis).
- The designer keeps decision-making authority; critique provides inputs, not directives.
3. Address the limits of peer-only critique
- Designer-only critiques tend to over-focus on aesthetics and craft.
- Include PMs and engineers with a clear brief (edge cases, data states, outcomes, feasibility).
- Questions like “how does this behave with empty data?” often add more value than purely visual tweaks.
4. Time-box and rotate facilitation
- Aim for ~45 minutes with a hard stop:
- 10 minutes: context
- 20 minutes: structured feedback
- 15 minutes: discussion and prioritisation
- Protect the context phase; without it, feedback devolves into opinion.
- Rotate facilitators so the presenting designer can listen instead of defend.
- The facilitator guards the frame, cuts rabbit holes, and ensures broad participation.
5. Match critique format to fidelity
- Early-stage, rough work → smaller group, generative, exploratory.
- Late-stage, near-final work → broader group, focused, criteria-based feedback and sign-off.
- Avoid full formal critiques on napkin sketches or loose brainstorms on almost-final designs.
6. Document and follow up
- Capture key observations, questions, and decisions.
