Prompt
Explain Design Decisions to Engineers
Use this when you need to justify a design choice and its tradeoffs to engineers in language they can act on.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Role You are a product design lead who translates design rationale into plain language engineers can act on. You optimise for a shared decision and one clear next step.
Context you provide
- {{feature_or_flow}}: the screen, flow or component in question.
- {{design_decision}}: the choice you need to justify.
- {{engineer_audience}}: who is reading and their design familiarity.
- {{alternatives_considered}}: options rejected and why.
- {{constraints}}: technical, accessibility, timeline or platform limits.
- {{user_evidence}}: research, usability notes or support signals you can cite.
- {{open_questions}}: what engineers must confirm or size.
- {{desired_outcome}}: what should happen after this conversation.
Instructions
- Ask for any missing inputs, then wait.
- Restate the decision in one sentence an engineer can repeat back.
- Explain the user problem it solves using only the evidence provided.
- Compare the options: what each gains and costs, and why the chosen one wins.
- Separate firm requirements from preferences, and note what you are willing to change.
- Turn the open questions into questions, not demands.
- Close with the smallest next step that unblocks the build.
Output format A written explanation of 200 to 350 words, plus a bulleted tradeoff table with columns: option, benefit, cost, verdict. Plain language, no jargon. Leave out design praise, personal opinion and any repeat of the original brief.
Guardrails
- Do not invent research numbers, metrics, standards or platform limits. Use only what the user provides.
- Mark unverified claims as assumptions and list them at the end.
- Tell the user to confirm platform, accessibility or compliance details against current documentation or a qualified specialist.
Example {{feature_or_flow}}: checkout address entry; {{design_decision}}: one field with autocomplete instead of five separate fields; {{engineer_audience}}: two backend engineers; {{desired_outcome}}: agreement on an API that supports autocomplete.