Most people treat ChatGPT like a search bar. They type a vague question, get a vague answer, then wonder why the output feels so thin. The instinct is to blame the tool. But the tool usually isn’t the problem. The prompt is.
I came across a brilliant breakdown from a LinkedIn creator who explained this perfectly, and it shifted how I think about every prompt I write. The author admitted they made this exact mistake for months when they first started using AI seriously. They’d type something like “tell me about marketing” and expect magic. What came back was filler they could’ve found on Google in 30 seconds.
Then this savvy professional changed one thing. They started treating prompts like briefs, the same way you’d brief a new hire before handing over a project. Specific. Structured. Loaded with context, constraints, and a clear outcome. The difference in output was immediate.
The big idea: a great prompt isn’t a question. It’s a brief. The more you treat the AI like a smart new hire who needs direction, the better your results get.
The 10 components of a perfect prompt
The original poster laid out 10 building blocks, and assigned each one a rough weight based on how much it moves the needle. I love this approach because it shows you where to spend your effort. Here’s the full breakdown, step by step, with the reasoning behind each one.
- Objective (15%): Define the task clearly. “Tell me about marketing” gets noise. “Create a 90-day marketing strategy for a SaaS startup targeting small businesses” gets results. This is your anchor, so make it sharp.
- Role (10%): Tell the model who to be. Senior software engineer. Expert copywriter. Startup advisor. Each role activates completely different reasoning patterns, so the same question gets answered through a different lens.
- Context (20%): Give background. Business type, industry, audience, goals, challenges, constraints. This is the heaviest-weighted piece for a reason. More context almost always means better output.
- Input Data (15%): Hand over the actual information it needs. Meeting notes, customer feedback, research reports. Never assume the model knows your situation, because it doesn’t until you tell it.
- Quality Checks (4%): Ask the model to verify its own work. Flag weak assumptions. Identify missing information. This kind of self-review catches a surprising number of errors before they reach you.
- Constraints (8%): Set limits. Max word count, no jargon, beginner-friendly, a specific budget range. Constraints force focus and make the output far more consistent.
- Examples (5%): Show it what good looks like. Give an input, the desired output, and the format. Examples cut down on misinterpretation in a big way.
- Iteration Request (5%): Tell it to improve. Ask it to critique its own response or suggest three alternatives. Prompting works best as a back-and-forth, not a one-shot.
- Instructions (10%): Tell it exactly what to do. Analyse. Identify. Prioritise. Strong action verbs do real work here and steer the model toward a concrete task.
- Output Format (8%): Specify the shape you want. Table, bullet points, step-by-step guide, JSON. Models follow structure reliably when you ask for it explicitly.
Why this framework clicks
What I appreciate most about the expert’s breakdown is the weighting. Notice that Context, Objective, and Input Data together make up half the formula. That tells you something important. The flashy stuff like roles and examples helps, but the real leverage is in being specific about what you want and feeding the model the raw material it needs.
This LinkedIn user also made a point that stuck with me. You don’t need every single component in every prompt. A quick question doesn’t need all 10. But for anything important, a strategy, a piece of writing, an analysis, layering in more of these pieces turns a generic response into something genuinely useful.
How to put it to work today
You don’t have to memorise all 10 at once. Here’s a simple way to start applying what the creator shared:
- Take a prompt you’d normally type in one line and rewrite it with a clear Objective and some Context first.
- Add a Role that matches the kind of expert you’d hire for the task.
- Paste in your real Input Data instead of describing it vaguely.
- End with the Output Format you actually want to read.
- If the first answer is close but not perfect, use an Iteration Request and ask for three improved versions.
That’s five of the ten components, and honestly it’ll already put you ahead of most people. Build the habit there, then layer in constraints, examples, and quality checks as you go.
The mindset shift is the whole game. Stop firing off one-line questions and hoping for brilliance. Start writing briefs. The mind behind this post proved the gap between average AI output and great AI output usually comes down to how much direction you give.
This is worth saving if you send a lot of prompts. Go check out the full LinkedIn post for the complete infographic with examples for each of the 10 components, and ask yourself where you land right now: structured prompt writer, or still winging it?