10-15 Minute Lesson Script
Introduction (2 minutes)
Welcome to Lesson 3! By now, you’ve had your first conversations with ChatGPT and you’ve learned the basics of prompt crafting. You understand the Four C’s, and you’ve probably noticed that some prompts work better than others.
Today, we’re taking it up a notch. We’re going to explore advanced prompting techniques that will make you feel like a pro—and we’re going to dive deeper into the art of iteration, which is really just a fancy word for “having a smart conversation with AI until you get exactly what you want.”
Think of today’s lesson as learning the difference between knowing how to drive and knowing how to drive well. You already know the basics. Now we’re going to show you the shortcuts, the techniques, and the little tricks that make everything smoother and more effective.
By the end of this lesson, you’ll know how to guide AI through complex tasks, get it to think step-by-step, and refine responses until they’re exactly right—without feeling frustrated or stuck.
Advanced Prompting Technique #1: Role Assignment (3 minutes)
Remember in Lesson 1 when we talked about system prompts? Let’s take that concept further with role assignment.
Role assignment is when you tell the AI to act as a specific type of expert or personality. This completely changes how it responds to you.
Here’s the magic: when you assign a role, the AI adjusts its tone, depth, and approach to match that role.
Basic prompt:
“Help me plan my garden.”
With role assignment:
“You are an experienced master gardener who specializes in teaching beginners. Help me plan my first vegetable garden. Explain everything in simple terms and warn me about common mistakes beginners make.”
See how much more specific that guidance becomes?
Here are some powerful roles you can assign:
For learning something new:
“You are a patient teacher who uses everyday analogies to explain complex topics.”
For creative work:
“You are an enthusiastic creative writing coach who gives constructive feedback and encouragement.”
For business tasks:
“You are a professional business consultant with 20 years of experience helping small business owners.”
For personal projects:
“You are my organized, detail-oriented assistant who helps me break big projects into manageable steps.”
The role you choose shapes everything that follows. Try different roles and see which voices work best for different types of tasks.
Advanced Prompting Technique #2: Step-by-Step Thinking (3 minutes)
Here’s a technique that dramatically improves AI responses, especially for complex problems: asking the AI to think step-by-step.
When you ask AI to “think step-by-step” or “show your reasoning,” it slows down and works through problems more carefully—kind of like when you talk through a problem out loud to figure it out.
Without step-by-step:
“How should I organize my home office?”
With step-by-step:
“I need to organize my home office. Think step-by-step about how to approach this. First, help me identify what categories of items I have. Then, suggest storage solutions for each category. Finally, give me a plan for maintaining the organization.”
This technique is especially powerful for:
- Problem-solving – “Walk me through this decision step-by-step”
- Planning – “Break this project into phases, explaining each one”
- Learning – “Teach me this concept step-by-step, checking that I understand before moving on”
- Decision-making – “Help me think through the pros and cons of each option”
You can even combine this with role assignment:
“You are my strategic thinking partner. I’m trying to decide whether to start a side business. Walk me through the key questions I should ask myself, step-by-step, before making this decision.”
Advanced Prompting Technique #3: Examples and Constraints (3 minutes)
The more specific you can be about what you want—and don’t want—the better your results will be. Two powerful ways to do this are showing examples and setting clear constraints.
Showing Examples:
Instead of describing what you want, sometimes it’s easier to just show the AI an example.
“Here’s an email I wrote that I really like: [paste example]. Now write a similar email about [new topic] in this same style.”
Or for creative work:
“Here’s a paragraph from my favorite author: [paste example]. I love how conversational and warm this feels. Write something similar about [your topic].”
The AI can analyze the style, tone, and structure and match it.
Setting Clear Constraints:
Constraints aren’t limitations—they’re guardrails that keep AI focused on what matters to you.
Useful constraints include:
- Length: “Keep it under 200 words” or “Write exactly 5 bullet points”
- Tone: “Professional but not stuffy” or “Encouraging and warm”
- Format: “Use short paragraphs, no more than 3 sentences each”
- Exclusions: “Don’t use jargon” or “Avoid mentioning competitors”
- Audience: “Write this for busy professionals who only have 2 minutes to read”
Powerful combined example:
“You are a marketing consultant. Write a 150-word introduction for my coaching services. Tone should be confident but approachable—like a friend who’s really good at what they do. Use short paragraphs (2-3 sentences max). Avoid corporate buzzwords. My audience is women over 40 who are considering a career change. Here’s a sample of my writing style: [paste example].”
The Art of Iteration: Having a Real Conversation (4 minutes)
Now let’s talk about iteration—which is really about treating AI like a collaborative partner, not a one-and-done tool.
Most people think they need to get the perfect prompt on the first try. But the truth is, the best results usually come from a conversation where you refine and adjust as you go.
Here’s what effective iteration looks like:
Round 1 – The Starting Point:
“Write a blog post introduction about time management for busy parents.”
AI gives you something, but it’s too formal.
Round 2 – Adjust the Tone:
“This is good, but it sounds too much like a textbook. Can you rewrite it in a more conversational, relatable tone? Like I’m talking to a friend over coffee.”
AI adjusts, but now it’s too casual.
Round 3 – Find the Balance:
“Better! But can you keep the warm tone while sounding a bit more professional? I want it to feel friendly but credible.”
Round 4 – Final Touch:
“Perfect! Now can you add one sentence at the end that hints at the 3 strategies I’ll share in the post?”
Result: Exactly what you needed—through conversation, not perfection.
Here are powerful iteration phrases to use:
For tone adjustments:
- “Make this more [formal/casual/friendly/professional]”
- “This sounds too [stiff/salesy/academic]—can you adjust?”
For length:
- “Cut this in half but keep the main points”
- “Expand on the second paragraph—give me more detail there”
For content:
- “I like everything except the ending—can you rewrite just that part?”
- “Add more specific examples”
- “Remove the technical jargon and explain it simply”
For alternatives:
- “Give me three different versions of this”
- “Try this again with a different approach”
Don’t be shy about asking for exactly what you want. The AI isn’t offended, isn’t tired, and actually gets better with clear feedback.
Combining Techniques for Power Prompts (2 minutes)
Here’s where it all comes together. The most powerful prompts combine multiple techniques:
Power Prompt Example:
“You are my experienced content strategy consultant who specializes in helping solopreneurs. I need help repurposing content.
I wrote this blog post: [paste or describe]
Think step-by-step about how I can turn this into:
- 3 social media posts (Instagram-length)
- 1 email newsletter introduction
- 5 tweet-sized tips
For each format:
- Match my conversational, encouraging tone
- Keep the core message but adjust for the platform
- Keep everything under [specific length]
- Make sure it speaks to busy entrepreneurs who need practical advice
Start with the social media posts.”
See how this combines: ✓ Role assignment (content strategy consultant)
✓ Step-by-step thinking (breaking down the task)
✓ Examples (the original blog post)
✓ Clear constraints (length, tone, audience)
✓ Structure (starting point to guide the AI)
Wrapping Up (1 minute)
Advanced prompting isn’t about making things complicated—it’s about being strategic. When you combine role assignment, step-by-step thinking, clear examples, and smart iteration, you turn AI from a basic tool into a genuine creative partner.
Remember: you don’t need to use every technique every time. Pick the ones that fit your task. And never forget that iteration is your friend—you can always refine, adjust, and improve.
This week, your mini challenges will give you hands-on practice with these techniques. By the end of the week, you’ll be crafting prompts that feel natural and get you exactly what you need.
Let’s make it happen!
3-5 Key Takeaways
- Role assignment transforms AI responses – Telling the AI to act as a specific expert (teacher, coach, consultant) changes its tone, depth, and approach to match your needs. Different roles work better for different tasks.
- Step-by-step prompts produce better results – When you ask AI to “think step-by-step” or “walk me through this,” it works through complex problems more carefully and gives you more thorough, logical responses.
- Examples are more powerful than descriptions – Instead of trying to describe the style you want, show the AI an example and ask it to match that style. This works for writing tone, format, and structure.
- Iteration beats perfection – The best results come from having a conversation with AI, not from crafting the perfect first prompt. Give feedback, ask for adjustments, and refine until you get exactly what you want.
- Combining techniques creates power prompts – Mix role assignment + step-by-step thinking + examples + constraints for sophisticated results. You don’t need all techniques every time—just pick what fits your task.
2-3 Practical Examples/Case Studies
Example 1: Janet’s Recipe Transformation
Background: Janet, 59, wanted to create a family cookbook as a gift but her handwritten recipes were incomplete and hard to follow.
Her approach:
Starting prompt:
“You are a professional recipe editor who helps home cooks write clear, easy-to-follow recipes. I’m going to give you my handwritten recipe notes, and I need you to turn them into a properly formatted recipe.”
First recipe she shared:
“Grandma’s pot roast – chuck roast, carrots, potatoes, onions, cook low and slow with some wine and broth. So good!”
AI’s response:
Created a complete recipe with ingredients list, measurements, step-by-step instructions, cooking times, and tips.
Janet’s iteration:
“This is perfect, but can you add a little note at the end? Something like ‘This was Grandma Rose’s Sunday dinner recipe—the house always smelled amazing when she made it.'”
Result: Janet converted 15 family recipes into a beautiful, professional cookbook by using role assignment and gentle iteration to add personal touches.
Key lesson: Role assignment (recipe editor) gave the AI the right framework, and iteration let Janet add the personal, family touches that made each recipe special.
Example 2: Robert’s Business Decision Framework
Background: Robert, 53, was trying to decide whether to expand his small consulting business by hiring his first employee—a big, scary decision.
His approach:
Power prompt combining multiple techniques:
“You are my strategic business advisor with experience helping small business owners make hiring decisions. I’m considering hiring my first employee for my consulting business.
Think step-by-step through the key questions I should ask myself before making this decision. For each question, explain:
- Why this question matters
- What answers would suggest I’m ready vs. not ready
- Any common mistakes people make at this stage
My context: I’m a solo consultant, current annual revenue is $120K, and I’m turning down work because I’m at capacity. The role would be an administrative assistant to handle client communications and scheduling.”
AI’s response:
Walked Robert through 8 critical questions: financial readiness, workflow systems, management skills, growth projections, and more.
Robert’s iteration:
“This is incredibly helpful. Can you now help me create a simple 3-month test plan? If I do hire someone, what milestones should I look for in months 1, 2, and 3 to know if it’s working?”
Result: Robert got a comprehensive decision framework AND an implementation plan, all from one strategic conversation.
Key lesson: Combining role assignment with step-by-step thinking helped Robert tackle a complex decision methodically instead of feeling overwhelmed.
Example 3: Elena’s Newsletter Voice
Background: Elena, 61, wanted to start a monthly newsletter for her art students but felt stuck—she didn’t know what her “newsletter voice” should be.
Her approach:
Using examples to find her style:
“I want to write a monthly newsletter for my watercolor painting students. I’m not sure what tone to use. Here are three samples from different newsletters I admire: [pasted three examples—one warm and chatty, one instructional, one inspirational].
Which style do you think would work best for art students who are hobbyists (not professionals), mostly age 45+, who take my classes for relaxation and creativity?”
AI’s response:
Analyzed all three, explained the pros and cons of each tone, and recommended the warm, chatty approach with gentle instruction woven in.
Elena’s iteration:
“That makes sense! Now write a sample 200-word introduction for my January newsletter. The topic is ‘Starting Fresh: New Year Projects for Watercolor Artists.’ Use that warm, chatty tone with gentle instruction. Here’s one of my class handouts so you can match my teaching style: [pasted example].”
Result: Elena got a newsletter draft that felt authentically like her, and now she has a template she uses every month.
Key lesson: Sometimes you don’t know what you want until you see examples. Using AI to analyze options and then create something based on those examples can help you discover your own style.
1 Hands-On Activity
Activity: “The Iteration Journey”
Time needed: 40-45 minutes
Objective: Practice all three advanced techniques (role assignment, step-by-step thinking, examples) and experience the power of iteration through a complete creative project.
Part 1: Choose Your Project (5 minutes)
Pick ONE project you’ll work on throughout this activity:
Option A: Write a personal introduction for a professional context (LinkedIn profile, website bio, speaker introduction, etc.)
Option B: Create a project plan for something you’ve been putting off (organizing a space, starting a hobby, planning an event, etc.)
Option C: Develop a piece of creative content (short story, blog post, newsletter, social media series, etc.)
Part 2: Round 1 – Role Assignment (10 minutes)
Start with a role-assigned prompt. Choose a role that fits your project:
For Option A (personal introduction):
“You are a professional brand consultant who helps people write authentic, compelling bios.”
For Option B (project plan):
“You are an organizational coach who specializes in helping people break overwhelming projects into simple steps.”
For Option C (creative content):
“You are a creative writing coach who helps people develop their unique voice.”
Now add your request. Example:
“You are a professional brand consultant who helps people write authentic, compelling bios. I need help writing a 150-word introduction for my LinkedIn profile. I’m a [your profession/role] who [what you do]. I want to sound [approachable/professional/creative—choose your tone].”
Run this prompt and save the response.
Part 3: Round 2 – Add Step-by-Step Guidance (10 minutes)
Now, ask the AI to help you improve what you got by thinking step-by-step:
“This is a good start. Now think step-by-step through how we can make this stronger:
- First, identify what’s working well
- Then, point out what could be improved
- Finally, suggest specific changes
Then rewrite it based on your analysis.”
Save this new version.
Part 4: Round 3 – Add Examples (10 minutes)
Now give the AI an example to match:
If you have something you’ve written before, paste it and say:
“Here’s something I wrote that feels authentically like me: [paste example]. Can you revise the bio to better match this voice and style?”
If you don’t have your own example, describe it:
“I want this to sound more like [describe: conversational/warm/straightforward/inspiring/etc.]. Think of [someone whose style you admire] but make it about my work. Rewrite it with that feel.”
Save this version.
Part 5: Final Refinement – Multiple Iterations (10 minutes)
Now do 2-3 quick iteration rounds to polish it exactly how you want:
Examples of refinement requests:
- “Make it 20 words shorter”
- “The second sentence feels awkward—can you rewrite just that part?”
- “Add one specific detail about [something important to you]”
- “Change the ending to be more [inviting/confident/curious]”
- “Give me two alternative versions of the opening sentence”
Keep refining until it feels just right.
Part 6: Reflect (5 minutes)
Look at all your saved versions (you should have at least 4-5).
Ask yourself:
- Which version is closest to what you wanted?
- What specific changes made the biggest difference?
- Which technique (role, step-by-step, examples, iteration) was most helpful?
- Did the final version surprise you in any way?
- Would you use this same approach for other projects?
Bonus: Save your final prompt (the one that got you the best result) as a template for future use!
3 Quiz Questions with Answers
Question 1
What is “role assignment” and when should you use it?
A) Assigning tasks to different AI tools
B) Telling the AI to act as a specific type of expert to shape how it responds
C) A way to organize your prompts into categories
D) Giving the AI permission to access certain information
Answer: B – Telling the AI to act as a specific type of expert to shape how it responds
Explanation: Role assignment is when you tell the AI to act as a specific expert or personality (like “You are a patient teacher” or “You are a business consultant”). This changes how the AI responds—its tone, depth, and approach all adjust to match that role. It’s one of the most powerful ways to customize AI responses for your specific needs.
Question 2
Why does asking AI to “think step-by-step” improve responses, especially for complex tasks?
A) It makes the AI work faster
B) It forces the AI to slow down and work through problems more carefully and logically
C) It’s required for the AI to understand your request
D) It makes responses shorter and simpler
Answer: B – It forces the AI to slow down and work through problems more carefully and logically
Explanation: When you ask AI to think step-by-step, it breaks down complex problems into smaller parts and works through them methodically—similar to how talking through a problem out loud helps humans think more clearly. This produces more thorough, logical, and useful responses, especially for planning, problem-solving, and decision-making tasks.
Question 3
Which statement about iteration is TRUE?
A) You should always get the perfect result on your first prompt
B) If you need to ask for changes, it means your prompt was wrong
C) The best results usually come from a conversation where you refine responses based on feedback
D) Iteration only works with advanced AI models
Answer: C – The best results usually come from a conversation where you refine responses based on feedback
Explanation: Iteration—having a back-and-forth conversation with AI where you give feedback and make adjustments—is actually how most professionals get their best results. It’s not a sign that you’re doing something wrong; it’s a smart strategy. You don’t need to nail the perfect prompt on the first try. Instead, start with a good prompt, then refine through conversation until you get exactly what you want. This works with any AI tool.