Why do some people make money with AI while others are still just “playing around”? (+ the C.L.E.A.R. framework)
While some complain that “AI writes rubbish”, others are already monetising these skills.
📊 Some real numbers (2024-2025):
• Freelancers with AI skills earn 40% more than their peers without those skills.
• 70% of freelancers say that thanks to AI they earn more than they did in their previous full-time job.
• AI saves a freelancer 8 hours a week on average (that’s a whole working day!).
• 75% of office workers already use AI in their work.
But why do some people get masterpieces while others get “hallucinations”?
❌ Typical mistake: we talk to AI as if it were Google.
We type “Analyse this data” or “Write a social media post” and hope for a miracle. But AI isn’t a magic bullet, it’s a talented intern. Without clear instructions, it simply guesses.
✅ The fix: the C.L.E.A.R. framework
This method turns your “intern” into a pro. Here’s how it works:
1. Context 🎭
Don’t just say “write some text”. Tell it who it is.
• The right way: “You are a social media marketing expert with 10 years’ experience creating viral content for tech companies.”
• The result: this changes the model’s whole approach to the task.
2. Limitations 🚧
Set the rules of the game.
• The right way: “LinkedIn post: up to 200 characters. Twitter: add emojis. Instagram: 5-7 hashtags.”
• Pro tip: you can use variables (from Google Sheets, for example) so you don’t have to rewrite the prompt every time.
3. Examples 💡
The secret ingredient everyone skips.
• The right way: show it something like “Here’s the input news story. Here’s what the perfect output post should look like.” This gives the model a better feel for tone and structure than any explanation could.
4. Actions 🎬
Spell out exactly what you want to get back.
• The right way: “Generate 3 posts and return them in JSON format.”
• Why: formats like JSON are ideal for automation (in n8n, for example), so the data can be passed along without errors.
5. Refinements (the safety net) 🛡
Foolproofing.
• The right way: “If the topic is controversial, keep a neutral tone. Always check the numbers before publishing.”
✴️The result?
Instead of chaos, you get perfectly formatted content: a professional tone for LinkedIn, a thread with a hook for Twitter and emojis for Instagram, all from a single prompt.
Your task for today:
Take one of your old prompts and add just the Context (the first part). You’ll be surprised how much the quality changes.
