Prompts for Designers: Briefs, Critiques, and Client Feedback

The most useful AI prompts for designers are the ones that help you work with client context: clarifying a vague brief, making sense of conflicting feedback, and explaining a design decision in terms a client will understand. A prompt that includes the approved direction, the audience, and the constraints produces a draft or critique you can use rather than one you have to rewrite. Tools like fenbs keep the client context and accepted decisions in one place, so each new session starts from what was actually agreed.

The templates below cover four things designers regularly need help with: extracting a clear brief from a client conversation, organizing feedback from multiple reviewers, explaining design choices to a non-designer, and structuring a design presentation.

What makes a designer prompt work

Design prompts fail when they ask an AI to make aesthetic decisions without the context that makes a decision right or wrong for this project. Before you send a design prompt, check that it includes these elements.

• The project goal: what success looks like for the client, not just what the deliverable is.
• The audience: who the design is for and what they need to do or feel.
• What is already decided: the brand direction, approved references, and rejected options.
• What is still open: the questions that need a decision before the design can move forward.
• The tone you need: explanatory, structured, strategic, or plain language for a non-designer.

Prompts for clarifying a brief

A brief clarification prompt turns a vague or incomplete client conversation into a working document with clear gaps. Run it before you start designing, not after you have made assumptions that need undoing. The questions it surfaces are worth asking before the first design rather than during a review.

I am starting a design project. Here are my notes from the client:

[paste notes or brief]

Help me write a one-page working brief that includes:
- The project goal in one sentence
- The primary audience and what they need from this design
- What success looks like for the client
- Any constraints I must work within (brand, format, deadline)

Then list the five questions I need the client to answer before I start.
Flag any contradiction in the notes.

Prompts for organizing client feedback

Client feedback often arrives as a mix of requests, reactions, and questions in the same email or call. Sorting that feedback into what is a decision, what is a question, and what is a preference makes it much easier to know what to act on and what to clarify before the next iteration.

Organize this client feedback on my design:

[paste the feedback]

Sort it into:
1. Direction changes I should act on now
2. Questions the client is asking (that I need to answer or ask them to decide)
3. Preferences that conflict with the original brief (flag these for discussion)
4. Comments that need no action

For each item in group 1, describe the change clearly enough that I can execute it.
For each item in group 3, explain why it conflicts.

For prompts that focus on the interface and user experience side of design work, prompts for UI and UX design covers user flows, accessibility reviews, and critique templates. For creating client-facing websites, prompts for website design has templates for the same context.

Prompts for explaining design choices

Designers often need to explain why a design decision is right for the project, not just describe what it looks like. A prompt that asks an AI to explain the reasoning behind a specific choice, in terms the client will understand, can help you prepare for a review conversation or write a clearer presentation.

Help me explain this design decision to a client who is not a designer.

Decision: [describe the design choice, e.g. using a single-column layout, a specific color, removing a button]
Why I made it: [your reasoning]
Project goal: [what the design is trying to achieve]
Client concern: [if the client has raised a worry or an alternative]

Write a short explanation (3-4 sentences) that:
- States the reason in terms of the project goal, not design theory
- Acknowledges the client's concern if there is one
- Ends with an open question if the decision still needs sign-off

Prompts for a design presentation

A design presentation that is organized around the client's goals rather than the design process lands better in a review. This prompt helps you structure the narrative for a presentation so the client understands why each decision was made before they are asked to approve it.

Help me plan the structure for a [design type] presentation to [client type].

Project goal: [what the design is trying to achieve]
Design decisions to present: [list the main choices the client needs to understand]
Client's priorities: [what they care most about]
Decisions they need to make today: [list any open questions or approvals needed]

Propose a presentation order that:
- Opens with the client's goal, not the design process
- Groups related decisions together
- Saves sign-off requests for after the rationale is clear

Keeping client context across iterations

Design projects involve many rounds of feedback, and the decisions from round one need to reach round three without a designer carrying them in their head. When an approved direction changes, the reason matters as much as the change. When feedback is rejected, the client's reason is worth recording so it does not come back in a later session.

After each client session, write the accepted changes, rejected ideas, and open questions into a record you can reference in the next prompt. That record makes it possible to start a new session or bring in a new team member without re-explaining the whole project history. Prompt engineering explains how to structure the context layer a reusable prompt needs.

The client context that matters most is often what was rejected. A designer who knows the client considered and ruled out a particular direction does not need to defend the chosen approach at the next review. A record that captures rejected ideas, with the client's reason, is as useful as one that captures approvals. It prevents a direction from cycling back when a new team member joins, when a stakeholder is replaced, or when an AI assistant generates suggestions that do not account for the project's history. Each rejected idea is a constraint on the next iteration, and a constraint with a reason is far more useful than one the team has to rediscover.

How to do this with fenbs

1. Create a free account and add the design project.
2. Write the approved brief and standing client constraints into Decisions and rules.
3. Save your most-used brief, feedback, and explanation prompts in Skills and prompts.
4. Connect your assistant over MCP from Integrations so it reads the client context before each session.
5. After each client review, update the decisions with what changed and add any new prompts that worked well.

fenbs does not import files from Figma or other design tools, and it does not generate images. It keeps the decisions and instructions your design process depends on.