On this page
Learn the four-part RCRQ formula (Role, Context, Request, Question) that turns AI from a search box into a conversation partner.
Key takeaways
- Google is a search engine. AI is a conversation partner. Talk to it like a colleague, not a search box.
- Think of AI as a 10-year-old genius: brilliant, but it needs clear instructions.
- Every great prompt has four parts: Role, Context, Request, Question (RCRQ).
- The secret weapon is one line: "Please ask me clarifying questions."
Do this now: Fill in this template with one thing you need help with, and send it to Claude.
ROLE: You are a world-class ___
CONTEXT: [Describe your situation — who you are, what's happening, any constraints or background info]
REQUEST: [What do you want AI to do? Be specific about format, length, tone, outcome]
QUESTION: Please ask me any clarifying questions before you begin.Class Notes
🧠 From Google Brain to AI Brain
You've spent years training your brain to talk to technology.
🙁 Short keywords.
💬 One-way communication.
🙏 Hope for the best.
That worked for Google.
It doesn't work for AI.
Google Brain:
- "best Italian restaurant near me"
- "how to fix printer wifi"
- "productivity tips 2026"
Short. Generic. No context.
You throw words at a search engine and hope something sticks.
AI Brain:
- "I need a romantic Italian restaurant for an anniversary dinner. We love authentic pasta, good wine, and an intimate atmosphere. Budget around $150 for two."
- "My HP OfficeJet Pro 8139e won't connect to WiFi. Blinking red light, high-pitched beeping. Already tried restart."
- "I'm overwhelmed with 3 client projects due this week, 5 team meetings to prep for, and quarterly planning. Help me build a focus game plan like Andrew Huberman would recommend."
Detailed. Conversational. Two-way dialogue.
The shift:
Google is a search engine.
AI is a conversation partner.
Talk to it like a colleague — not a text box.
🧒 Your AI Partner: A 10-Year-Old Genius
Here's the mental model you need:
→ Think of AI like a 10-year-old genius who knows everything about everything.
What does that mean?
AI is brilliant.
They have access to more information than you could ever imagine.
→ But they need super clear instructions to give you what you actually want.
Just like a real 10-year-old, they need:
- Details and context — not just "clean your room" but exactly what "clean" looks like
- To understand WHY — what's the goal behind the request?
- Examples — show them what good looks like
- Patience — they'll make mistakes, and that's part of the conversation
- Permission to ask questions — so they can fill in what's missing
▶️ See the Difference: Try It Right Now
Step 1: The Google Brain approach
Open ChatGPT or Claude.
Type this exactly:
Tell me a good restaurant in my area.Look at the result.
Generic? Vague? Asking you a bunch of follow-up questions?
Step 2: The AI Brain approach
Now try this — but make it YOURS:
You are Jonathan Gold, Pulitzer Prize–winning food critic and expert at uncovering intimate, authentic dining experiences.
I’m planning a date night this Saturday in [CITY].
We love incredible Italian food, especially homemade pasta and a strong wine list.
We want a cozy, intimate atmosphere — not loud, not trendy, not rushed.
Budget is around $150 total for two people (food + wine).
This is a romantic date, so ambiance matters as much as the food.
REQUEST:
Recommend 2 Italian restaurants that best match this exact vibe.
For each recommendation, include:
- 2 sentances on why it’s ideal for a date night
- Noise level and seating intimacy
- Reservation tips for a Saturday night
QUESTION:
Please ask any clarifying questions
Step 3: Compare
- Which response was more useful?
- Which one felt like AI actually understood what you wanted?
- What details made the difference?
This is the shift.
Same AI. Same question.
Completely different results.
The difference is how YOU communicate.
✍️ The RCRQ Framework — Your Prompt Structure
Now that you've felt the difference... let's give you a structure to make this repeatable.
When you sit down to type... \what do you actually say?**
→ Write this down: R - C - R - Q
⚡️ The 4 Parts of Every Great Prompt:
| Letter | Stands For | What To Include |
|---|---|---|
| R | Role | Who is AI in this conversation? "You are a world-class ___" |
| C | Context | What's the situation? Background, details, constraints |
| R | Request | What do you want? Be specific about the outcome |
| Q | Question | "Please ask me clarifying questions" |
‼️ Why Each Part Matters:
- Role → Focuses AI's expertise (marketing coach vs. therapist vs. chef)
- Context → Gives AI what it needs to help YOU specifically — not generic advice
- Request → Clear outcome, not vague hope
- Question → THE SECRET WEAPON — lets AI slow down and fill in gaps
🔑 The Secret Weapon: "Please Ask Me Clarifying Questions"
This one line changes everything.
Why?
AI wants to help fast. If you don't give it permission to ask, it'll just guess.
Without it: AI rushes to an answer with missing information.
With it: AI slows down, asks what it needs, and delivers exactly what you want.
This one line turns a vending machine into a thinking partner.
🧱 Example: See RCRQ in Action
Let's say you need help writing an email about a project delay.
👎 Without RCRQ:
Help me write an email to my team about the project delay.Generic. No context. AI doesn't know your tone, your team, or the situation.
😭 You'll get a bland, corporate-sounding template that doesn't fit.
👍With RCRQ:
ROLE: You are an expert in team communication and leadership messaging.
CONTEXT: I'm a marketing director at a SaaS company. We just found out our product launch is delayed by 3 weeks due to a vendor issue. My team has been working hard and I don't want them to feel defeated. This is my first time delivering bad news as a leader.
REQUEST: Help me write an email to my team that acknowledges their hard work, explains the delay honestly without blaming anyone, and refocuses us on the adjusted timeline. Keep it under 200 words and make it feel human, not corporate.
QUESTION: Please ask me any clarifying questions before you write.🏎️ What Happens Next:
AI asks:
"How big is your team?”
“What's the company culture like — casual or formal?”
“Is there a silver lining you can share?"
You answer.
AI writes an email that actually sounds like YOU.
Same task. 10x better result.
Reminder:
You don't have to label each section "ROLE:" or "CONTEXT:" — AI understands natural language.
But when you're learning, the labels help you make sure you're not missing anything.
RCRQ is your checklist until it becomes second nature.
▶️ Your Turn: Build Your First RCRQ Prompt
Let's put this into practice right now.
🤔 Step 1: Think of ONE thing you need help with.
- A task you've been putting off
- A message you need to write
- A decision you're stuck on
- Something you'd normally Google 10 times
👨💻 Step 2: Fill in this template:
ROLE: You are a world-class ___
CONTEXT: [Describe your situation — who you are, what's happening, any constraints or background info]
REQUEST: [What do you want AI to do? Be specific about format, length, tone, outcome]
QUESTION: Please ask me any clarifying questions before you begin.💬 Step 3: Paste it into ChatGPT or Claude
✅ Step 4: Answer AI's clarifying questions
🦄 Step 5: Watch the magic
🔥 Hot Tip: You don't have to type all of this.
If typing long context feels overwhelming — talk instead.
Wispr Flow lets you speak and it types for you — anywhere on your computer.
Brain dump your context out loud. AI doesn't care if it's messy.
We'll cover Wispr Flow in the Tools & Downloads section.
🎉 What’s Next
Next, you’ll stop asking, “What should I type?”
You’ll start with what’s bothering you: Pain → Problem → Prompt.
And you’ll use AI to build the prompt for you—so you never stare at a blank page again.
⬅️ Back to your pre-watch list: Helping Hands (all videos) · General · Advanced
