Dear Techies,

It's not the tool. It's the question.

Most people try an AI tool, get a mediocre response, and quietly conclude that AI is overhyped.

They're not wrong about the response.

They're wrong about the cause.

Where the frustration actually comes from

AI tools like ChatGPT, Claude, and Gemini are not search engines. They don't look things up. They predict the most useful response based on what you give them.

Which means the quality of what you get out is almost entirely determined by the quality of what you put in.

Vague prompt. Vague answer. Every single time.

The people getting genuinely impressive results from AI aren't using better tools or paying for premium subscriptions. They're just asking differently. And there's a simple framework behind how they do it.

The RCTF framework — four parts, dramatically better results

You don't need all four components every time — but the more you include, the better your results.

  • Role: Tell the AI who to be.

    For instance, "Act as a plain-English explainer" or "respond as an experienced project manager" completely changes the tone, vocabulary, and approach of everything that follows. Think of it as briefing someone before they start working for you.

  • Context: Give it the background it needs.

    Who is this for? What's the current situation you are solving for? What have you already tried? Most people skip this entirely — and it shows in their results.

  • Task: Be specific about what you actually want to get done.

    "Help me with my presentation" is a task but "Write an opening slide for a ten-minute presentation on climate change for secondary school students, starting with a surprising statistic" is a useful task. There's a big difference.

  • Format: Tell it how to deliver the answer.

    For instance, a numbered list, three short paragraphs, under 150 words, no jargon, etc. Without this, you get a wall of text by default. Even when that's the last thing you need.

Role. Context. Task. Format.

That's the whole framework.

The part most people miss

Getting a decent first response is only half of it.

The biggest difference between people who find AI genuinely useful and people who don't isn't the prompt. It's what they do next.

Prompting is a conversation, not a one-shot command. The first response is always a draft so treat it like one. Refine it. Push back. Ask for a shorter version, a warmer tone, a completely different angle. Here are some useful phrases you can use:

"Go deeper on point two."

"Rewrite this for someone who knows nothing about the topic."

"Give me three alternatives."

That back-and-forth is where the real value lives. And it takes about thirty seconds.

This week's video

The new video walks through the full RCTF framework with live examples — a professional email, a complex topic explained simply, and a structured launch plan, so you can see exactly how each component changes the output in real time.

It's under fifteen minutes (1.5x). No technical background needed. And you can apply what you learn the same day.

👉 Watch it here: Video Link

Three things to try this week

  1. Pick one task you'd normally work through yourself for instance, an email, a summary, a plan, and build a prompt using all four parts of the framework

  2. After the first response, refine it at least once before you use it and notice how quickly it improves

  3. Pay attention to which part of the framework made the biggest difference because it varies from person to person and task to task

Before you go

If you know someone who's tried AI tools and quietly given up because the results weren't good enough, this is the video to send them.

The framework is simple enough to use immediately and makes a noticeable difference from the very first try.

Stay Savvy,
The Tech Savvy Starts Here Team

P.S. This works with any AI tool — ChatGPT, Claude, Gemini, Copilot, whatever you're already using. You don't need to switch anything. You just need to ask differently.

Enjoyed this edition?
Forward it to a friend or colleague who will enjoy it as well.
Missed something? Catch up in the newsletter archive.

🧠 Keep learning. | 💬 Keep questioning. | 💥 Keep growing.

Keep Reading