5.2 Be Explicit — No Mind Reading

State your requirements clearly.

🎯 Core Goals

  • Learn why “more is more” when it comes to prompt instructions.
  • See how specificity leads to dramatically better results.

If you don’t specify it, the LLM will guess — and it will guess based on the most common pattern it’s seen, not what you actually need. The more explicit you are, the less room there is for a wrong assumption.

👁️ Visuals & Interactives

The Specificity Stack

Click the layers to turn a vague prompt into a great one.

Prompt Preview
"Write an email."
AI Result Quality
Poor (Generic)

📝 Key Concepts

  • Can you specify what you want?
  • Can you describe in detail what you do for a living?
  • Does the answer you expect from the LLM have a clear right or wrong answer?
  • Can you improve the prompt above?

The 6W Framework for better prompts: When stuck on why your result isn’t right, ask yourself if you covered: Who is the audience? What exactly do you want? When/Where is the context? Why does this matter? How should it be formatted? More answered Ws = better output.

For LLM-assisted coding or project management, it’s very important to have a clearly defined specification. We’ll explore this when we cover project management.

How do you improve a bad prompt? Add more context. The most common reason an LLM gives a generic or off-target response is that the prompt was ambiguous. Read your own prompt as if you knew nothing about the task — would the instruction be clear to a stranger? If not, add more detail.

arrow_back Next arrow_forward