A couple of years ago, I was one of those people who tried AI, decided it would take more time to figure out than it was worth, and went back to doing things myself.
The tool wasn't bad, I was just terrible at using it.
I’d type something vague, get something useless back, spend ten minutes regretting it, and close the tab. The problem wasn’t the AI. The problem was that I was talking to it like a search engine, dropping a few keywords and hoping for something useful. I’ve heard people call this Google Brain, and it’s the fastest way to get garbage output from a genuinely powerful tool.
The shift happened when I started treating AI less like a search bar and more like a very capable colleague who needs a proper briefing before they can do anything useful. You wouldn’t walk up to a smart teammate, say “write me something about productivity,” and expect something good back. You’d sit down with them, explain the situation, tell them what you’ve already tried, describe what you actually need, and let them ask a few questions before they got started. That’s the whole game.
The Framework That Changed Everything
The framework I use now—one I first came across through AI strategist Callan Faulkner—is RCRQ: Role, Context, Request, Question.
Role
Tell AI who it’s supposed to be. “You are a world-class product strategist.” “You are an expert in time management.” It sets the lens through which everything else gets read.
Context
The full situation: what you’re working on, what you’ve already tried, what’s at stake, who the audience is. The briefing you’d give a smart colleague on their first day.
Request
The specific deliverable, the format, the length, the tone, what you actually want back. Be precise about the output you need.
Question
The line you add at the end of every real prompt: “Please ask me clarifying questions before you answer.” This one line changes everything.
That last element—the Question—is the one most people skip. Ending your prompt with “please ask me clarifying questions before you answer” slows AI down, stops it from rushing to a confident wrong answer, and surfaces questions you probably should have been asking yourself anyway.
You wouldn’t walk up to a smart teammate, say “write me something about productivity,” and expect something good back. That’s the whole game.
The First Answer Is Never the Answer
The other shift worth naming is that the first answer is never the answer—it’s the start. The people who get the most out of these tools are the ones who keep the conversation going, push back, redirect, ask for a different angle, and keep refining from there.
If you’re in there using AI every day, even if the results aren’t great yet, you’re already ahead of most people. Early friction isn’t evidence that the tool isn’t worth learning. It’s just the learning part.
Where the Human Still Matters
If you’re wondering where the human goes in all of this, here’s how I think about it. AI gets me to sixty or seventy percent. It organizes my thinking, synthesizes information, lays down the foundation. The remaining thirty to forty percent is me: the critical thinking, the creativity, the judgment call that requires context AI will never have.
That human layer isn’t going away, and it shouldn’t. The goal is to stop spending my best thinking on organizing details, and save it for the work that actually needs me.
That’s what a good prompt is for.