🧭 AI Library - Writing Prompts: 1-2-3!
             
  Ask → Review → Refine 

Golden Rule

If other people can’t understand what you’re asking, the AI probably won’t either.

Ask → Review → Refine : A Simple Method for Better Questions
This simple, three-step method helps anyone learn to write better prompts, achieve better results, and develop effective follow-up questions. Once mastered, it makes advanced prompting techniques easier to learn.
🔥 Need help phrasing your question? Ask the AI!

1️⃣ Ask A Question

What to do: Start with a clear, natural-language question. Be specific about what you want the AI to do — explain, summarize, compare, create, or analyze.

Example: “Explain the difference between revenue and profit.”

Tip: Need help phrasing your question? Ask the AI for help!

2️⃣ Review The Answer

What to do: Read the AI’s answer critically. Ask yourself:

  • Does it make sense?
  • Does it answer what I meant?
  • Is anything missing, vague, or off-topic?

Example: “That’s helpful, but can you include an example from a real company?”

Goal: Develop the habit of evaluating AI responses — not just accepting them.

3️⃣ Refine The Question

What to do: Revise your question based on what you learned. Add details, constraints, or clearer intent — then ask again.

Example: “Compare Apple’s and Samsung’s profit margins in the last fiscal year.”

Each refinement brings the AI closer to what you really need — that’s iteration.

💡 Why It Works

This three-step loop builds the same habits expert users apply:

  • Clarity of intent (Step 1)
  • Critical thinking (Step 2)
  • Iterative improvement (Step 3)

Once comfortable with this rhythm, you can explore advanced techniques like chain-of-thought reasoning, persona prompts, and contextual framing.

🧠 Try It Yourself

Write your own prompt below and experiment with the Ask → Review → Refine process. Copy your text directly into ChatGPT to test it.

🟠 The Golden Rule: If other people can’t understand what you’re asking, the AI probably won’t either.

Written by ChatGPT | Edited by Peter Z. McKay | © 2025 UF Business Library AI Series