What Makes a Good Prompt? A Simple Framework for Better AI Conversations
If AI outputs always feel generic and off-target, the tool isn't the problem — the way you're asking is. Learn a simple three-part framework for writing prompts that actually get you useful results.
What Makes a Good Prompt? A Simple Framework for Better AI Conversations
You have an important email to write. You turn to ChatGPT for help. You type: "Write me an email." The output arrives — generic, flat, and nothing like what you had in mind. Frustration sets in, and you think: "This AI is useless."
But the tool isn't the problem. The way you're asking is.
When AI Feels Like It's Not Working
This feeling is incredibly common — and completely understandable. We're used to working with search engines: type a few words, get results. But AI isn't a search engine. It's more like a smart colleague who knows absolutely nothing about what's in your head — unless you tell them.
Ask a vague question, get a vague answer. That's a simple rule, not a flaw in the technology.
The Difference Between a Weak Prompt and an Effective One
Let's look at two real examples side by side.
Example one — writing an email:
Weak prompt: "Write me an email" Output: A generic email that's probably not usable.
Effective prompt: "You are a professional assistant. I need to write a formal email to a client confirming our meeting next week. The tone should be warm but professional. Keep it to five sentences or fewer." Output: Exactly what you were looking for.
Example two — summarizing a document:
Weak prompt: "Summarize this text" Output: A summary that might be too long, too short, or focused on the wrong things.
Effective prompt: "You are an analytical assistant. Summarize this report for a busy manager who has limited time. Give me only the three key takeaways as bullet points." Output: Three clear, actionable points.
The difference is significant — and it came from just a few extra sentences.
A Simple Framework: Three Elements of an Effective Prompt
You don't need a specialized course. A three-part structure is all it takes:
1. Role — Tell AI who to be. Example: "You are a marketing consultant."
2. Context — Give it the information it needs. Example: "I run an online children's clothing store and I'm looking to attract new customers."
3. Expected output — Be specific about what you want. Example: "Suggest five practical Instagram post ideas that target young mothers."
These three elements noticeably change the quality of what you get back. Not because AI suddenly became "smarter" — but because it now has enough information to actually help.
A Common Mistake Worth Avoiding
A lot of people write a prompt once, look at the output, and if it's not quite right, they put the whole tool down.
But prompting is a conversation, not a one-way command.
If the first output isn't exactly what you needed, keep going: "That was good, but the tone is too formal — rewrite it in a friendlier voice." That one sentence can completely change the result. Refining and rewriting your prompt is a natural part of the process, not a sign that something went wrong.
Where to Start
Pick one recurring task from your week — an email, a summary, a message — and rewrite it using the three-part framework. Just that one exercise will show you the difference.
If you'd like to practice this with real examples from your own work, we can start exactly where you are in a free 20-minute consultation.