In partnership with
You Don’t Need to Be Technical. Just Informed
AI isn’t optional anymore—but coding isn’t required.
The AI Report gives business leaders the edge with daily insights, use cases, and implementation guides across ops, sales, and strategy.
Trusted by professionals at Google, OpenAI, and Microsoft.
👉 Get the newsletter and make smarter AI decisions.
Most people treat AI like a magic 8-ball. They throw in vague requests and hope for the best. “Help me write a blog post” gets you generic fluff. “Make this better” gets you confused, rambling.
What if I told you that you can beat 98% of people using AI with just two simple systems?
There's a reason some people consistently get AI to produce gold while others get digital garbage. They're using structured frameworks instead of shooting from the hip.
Today, I'm sharing the two prompting systems that separate AI pros from AI novices. After testing them for two weeks, these frameworks consistently put out results that make people ask, “How did you get AI to do that?"
Why Most People Fail at AI (And How These Frameworks Fix It)
The biggest mistake? Treating AI like Google search. You wouldn't ask a junior employee to "make the website better" without context. Yet that's exactly how most people prompt AI.
The typical failure pattern:
Vague task → Generic output → Frustration → "AI isn't that useful"
The 2% approach:
Structured input → Specific context → Targeted output → Consistent value
The difference isn't the AI model. It's the human operating it. These frameworks turn your scattered thoughts into AI instructions that actually work.
Framework #1: The TCREI Method for Bulletproof Prompts
TCREI stands for: Task, Context, Resources, Evaluate, Iterate
Here's how it works:
T - Task: Define exactly what you want AI to do. Not "help me write" but "create 3 LinkedIn posts promoting my new course on Excel automation."
C - Context: Feed the background information. Your company details, target audience, brand voice, launch date. More context = better results.
R - Resources: Provide examples of what good looks like. Screenshots, competitor posts, previous content that worked. Show, don't just tell.
E - Evaluate: Read the output like a skeptical customer. Does it hit the mark? What's missing? What's wrong?
I - Iterate: Refine based on your evaluation. "Make the tone more conversational" or "add specific pricing details."
Copy-paste example:
TASK: Create an Instagram post marketing my new productivity app called FocusFlow.
CONTEXT: I'm targeting remote workers aged 25-40 who struggle with distractions. The app launches Monday. My brand voice is friendly but professional, like a helpful coworker.
RESOURCES: Here's my most successful post from last month [paste example]. And here's a competitor post I admire [paste example].
OUTPUT FORMAT: Start with a hook question, include 2-3 key benefits, end with soft CTA and 3 relevant hashtags.This beats "create an Instagram post for my app" by miles.
Framework #2: The RSTI System for When Results Still Suck
RSTI stands for: Revisit, Separate, Try different phrasing, Introduce constraints
Use this when TCREI gives you 80% but you need 95%.
R - Revisit: Go back through your first framework. Can you add more context? Better examples? Remove conflicting information?
S - Separate: Break long prompts into shorter sentences. Instead of one paragraph describing everything, use bullet points or numbered steps.
T - Try different phrasing: Reframe the task entirely. Instead of "write a speech," try "write a compelling story that illustrates [your main point]."
I - Introduce constraints: Add specific limitations to force better results. "Only use words a 7th grader would understand" or "Include exactly 3 actionable tips."
Real example:
ORIGINAL: Help me write a newsletter about time management that's engaging and useful.
RSTI VERSION:
TASK: Write a story about someone who went from chaotic to organized in 30 days.
CONSTRAINTS:
- Use only simple words (7th grade level)
- Include exactly 3 specific tactics they used
- End with one action readers can take today
- Maximum 400 wordsThe second version forces AI to be specific, accessible, and actionable.
Your Implementation Game Plan
Pick one thing you've been struggling to get AI to help with. A blog post, email sequence, social media content—anything.
Here's your testing sequence:
First attempt: Use TCREI on your chosen task. Spend 5 minutes on context and resources.
If needed: Apply RSTI to refine your results when they're close but not quite right.
Save for later: Document your best prompts in a "Templates" file for future reuse.
The goal isn't perfection on attempt one. It's building the habit of structured prompting that compounds over time.
Most people will read this and do nothing. The 2% will spend 10 minutes testing these frameworks and immediately see the difference.
Which group are you in?
Keep shipping,
Luke