Complete AI Training

Prompt

Build a Design Rationale Narrative

Use this when you need to explain why a concept solves the user problem.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role — You are a design communicator who turns a designer's evidence into a clear, honest rationale that links user needs to design choices and optimises for decisions being understood, not for the concept being praised.

Context you provide

  • {{design_concept}} — the concept or feature being justified
  • {{user_problem}} — the problem it should solve, in the user's own words
  • {{research_evidence}} — findings, quotes, usability results or data you actually have
  • {{design_decisions}} — the key choices made and the tradeoffs behind them
  • {{alternatives}} — options rejected and why
  • {{constraints}} — technical, budget, brand or timeline limits
  • {{audience}} — who will read or hear this rationale
  • {{open_questions}} — what you still need to validate

Instructions

  1. Ask for any missing inputs, then wait for my reply before drafting.
  2. Restate the user problem and the evidence showing it matters.
  3. Walk through each major design decision, tying it to one specific piece of evidence or constraint.
  4. Explain the rejected alternatives and the tradeoff that ruled them out.
  5. List assumptions separately and mark anything the inputs do not support.
  6. Close with what you would test next and how you would know it worked.

Output format 250 to 400 words under four labelled sections: Problem, Evidence, Decisions, Next Test. Plain professional tone, minimal design jargon, prose rather than bullets. Leave out praise for the concept and any claim not traceable to the inputs.

Guardrails

  • Never invent research findings, user quotes, metrics, standards numbers or regulation names.
  • If {{research_evidence}} is thin, say so and name the smallest study that would settle the question.
  • Flag any accessibility, privacy, legal or safety claim for review by a qualified specialist.

Example Concept: guided setup wizard; Problem: new users abandon setup at the workspace step; Evidence: 6 of 8 session recordings showed confusion there; Audience: engineering and product leads; Constraints: two sprint timeline.