3 Prompt Mastery – Experiment and Iterate (5 mins)

Module 3: Experimentation – How to Iterate for Better Responses

Intro:
Getting great AI responses isn’t about luck—it’s about iteration. In this module, you’ll learn how to systematically tweak your prompts to get more accurate and useful results.

Learning Objectives:

  • Understand why prompting is an iterative process
  • Learn how small changes in wording affect AI-generated responses
  • Use role-playing and specificity to guide AI’s output
  • Practise improving a basic prompt through three iterations

written version of the video tutorial

Getting great AI responses isn’t about luck—it’s about iteration. The best AI users know that small tweaks to their prompts can dramatically improve the quality of responses.

In this module, you’ll learn:

  • Why prompting is an iterative process, not a one-time action
  • How small wording changes affect AI-generated responses
  • How role-playing and specificity can improve AI’s accuracy
  • A simple three-step process to refine and test prompts
  • A practical exercise to improve a basic prompt in three iterations

By the end of this lesson, you’ll know how to adjust your prompts systematically to get the best possible AI-generated answers.

Why Prompting is an Iterative Process

Many people assume that AI should generate perfect answers immediately—but that’s not how it works.

AI models don’t “think” like humans. They predict words based on patterns, which means responses can be too vague, too broad, or not quite what you need.

Instead of accepting the first response as final, treat prompting as an interactive process. Think of AI like a junior assistant: it gives a first draft, and you refine the request until you get exactly what you need.

Each time you tweak the prompt, you guide AI closer to the best result.

The Small Changes, Big Impact Principle

Even minor changes in wording can significantly alter AI’s response.

Here’s an example:

❌ Weak Prompt: “Give me advice on productivity.”

  • The AI may return generic tips that aren’t specific to your situation.

✅ Improved Prompt: “Give me five productivity strategies for a remote worker who struggles with distractions.”

  • Now, the AI has a clearer focus and will provide more relevant advice.

Let’s refine it even further:

✅ Even Better Prompt: “Give me five productivity strategies for a remote worker who struggles with distractions. Keep each tip under 20 words.”

  • This controls the response format, making it concise and actionable.

By making small adjustments, we dramatically improve the response.

Using Role-Playing and Specificity to Improve AI Responses

Another powerful way to refine AI outputs is to use role-playing and specific context in your prompts.

For example, instead of:
❌ “Explain machine learning to me.”

Try:
✅ “You are an expert AI researcher. Explain machine learning to me as if I’m 12 years old, using simple analogies.”

Or, if you want a professional answer:
✅ “You are a university professor. Give me a two-paragraph explanation of machine learning for a beginner.”

By defining who AI should act as and who the response is for, you get tailored, more relevant answers.

The Three-Step Process for Refining Prompts

If your AI response isn’t useful, don’t assume AI is bad—refine your prompt instead.

Step 1: Identify What’s Missing

Ask yourself:

  • Is the response too vague?
  • Is it too broad or too long?
  • Does it lack structure?

Step 2: Add Clarity, Context, or Constraints

  • Be more specific about the topic or desired response.
  • Provide more context (who, what, where, why).
  • Control the format (bullets, structured lists, word limits).

Step 3: Compare & Adjust

  • Run the improved prompt and see if it delivers a better response.
  • Make additional tweaks if needed.
  • Save the best version for future use.

Practical Exercise – Improve a Prompt in Three Iterations

Try this step-by-step approach:

1️⃣ Basic Prompt: “Tell me about AI ethics.”
2️⃣ Refined Prompt: “Give me a short, three-bullet summary of AI ethics, with real-world business examples.”
3️⃣ Final Prompt: “Summarise the three biggest AI ethics concerns in business today, with an example for each. Keep each answer under 30 words.”

Compare the AI responses at each stage—see how they improve with each refinement.

Now, try this with your own prompt!

Summary & Next Steps

Key Takeaways:

✔️ AI prompting is an iterative process—tweaking your prompts leads to better responses.
✔️ Small wording changes can dramatically improve output.
✔️ Role-playing and specificity help AI tailor its response.
✔️ A simple three-step refinement process helps improve any AI-generated response.

📌 Action Step: Take a recent AI prompt you’ve used and improve it using the three-step process. Notice how each small change improves the response.

🚀 Next Module: Now that you know how to refine AI responses, we’ll explore how to store and organise your best prompts so you can reuse them easily. Let’s dive in!

Previous Lesson
Back to Course
Next Lesson