Master the Role + Target + Format Prompting Technique to Get Better AI Answers

[Intro music fades in, then out] Hey, it’s Mal, your Misfit Master of AI, and this is “I Am GPTed” — the show where we make ChatGPT, Claude, Gemini, Grok, and the rest of the robot squad slightly less useless and a lot more helpful. Let’s get right into it before the hype bros show up with a 40-slide AI keynote. --- So, one specific prompting technique that actually moves the needle: **Role + Target + Format**. You’re not just asking the AI for stuff; you’re casting it in a role, telling it who it’s talking to, and how you want the answer. Here’s the “before” — the classic rookie move: > “Explain blockchain.” And then you wonder why you get a textbook mixed with a sleep aid. Now the “after”: > “You are a high school teacher who hates jargon. Explain blockchain to a 15-year-old who likes online games. Use short sentences and give me 3 bullet point examples.” Same question, completely different brainpower. One feels like homework, the other feels like someone is actually trying to help you not feel dumb. Use this across ChatGPT, Claude, Gemini, Grok — they all perk up when you stop treating them like a search bar and start treating them like interns with job descriptions. --- Let’s talk **practical use case** you probably haven’t tried: **decision drafts**. Not “write my essay” or “help with email” — I mean: “Help me decide like a functioning adult.” Example: > “You are a pragmatic career coach. I’m choosing between two job offers. Lay out a simple comparison table: salary, commute, stress level, growth potential, and ‘how likely I am to hate my life in 6 months.’ Then give me 3 questions I should ask myself before deciding.” That’s ChatGPT or Claude as your reality-check friend — without the side order of judgment. You can use the same trick for choosing software, vacation plans, even whether to renew that subscription you forgot you had. --- Now, **common beginner mistake** time — and yes, this is one I made loudly and repeatedly: Treating AI like Google. I used to type stuff like: > “Best productivity tips.” Then I’d sit there reading a bland list that looked like every blog post ever written, thinking, “Wow, AI is overrated.” The problem wasn’t the AI. It was me being vague. The fix is context. Instead of that, say: > “I’m a project manager working remotely, constantly in meetings, with two kids under 6. Give me 5 realistic productivity tips I can actually start this week, with one sentence on how to implement each.” Suddenly the answer sounds like it was written for a human with an actual life, not a robot monk in a cave. So if you’ve done this, congrats: you’re repeating my early mistakes. You’re in terrible but familiar company. --- Here’s a **simple exercise** to build your AI interaction muscles: For your next three prompts, always include three things: 1. “Act as a…” — give it a role. 2. “For [who]…” — define the audience. 3. “In [format]…” — tell it how to package the answer. For example: > “Act as a friendly tutor. Explain basic budgeting for a 25-year-old who’s never managed money before, in 5 bullet points.” Then: > “Act as a skeptical editor. Rewrite this paragraph to be clearer and shorter, and tell me what was confusing.” Do that three times today. You don’t need a course. You need reps. --- Last piece: **how to evaluate and improve AI-generated content** without needing a PhD or a therapist. Use what I call the **Three-Question Check**: 1. **Does this sound like me?** If it sounds like a corporate press release or a robot on LinkedIn, tell it: > “Rewrite this in a more casual, human tone that sounds like a real person talking, not a PR department.” 2. **Is anything obviously wrong or vague?** Ask: > “Highlight any claims that need sources or examples. Then add one concrete example to each.” 3. **Is it actually useful?** Ask: > “Turn this into a checklist or step-by-step guide I can follow in under 10 minutes.” Treat every AI answer as a **first draft**, not divine wisdom. The AI types fast. You do the steering. --- All right, that’s it for today’s dose of “I Am GPTed” with Mal, your Misfit Master of AI. If this helped you level up your prompts — or at least convinced you to stop typing “write me something good” — make sure you **subscribe to the podcast** so you don’t miss future episodes. **Thanks for listening**, seriously. You could be doomscrolling, but you chose to level up instead. This has been a **Quiet Please** production. To learn more, head over to **quietplease dot ai**. [Outro music fades out] For more check out https://www.quietperiodplease.com/ and for some great deals go to https://amzn.to/4nidg0P