Did ChatGPT provide you with an ideal reply? Use this immediate to seek out out why

I’ve written earlier than about find out how to ask ChatGPT to enhance a so-so immediate. However what about these occasions with ChatGPT, Claude, or Gemini completely hits the nail on the top, supplying you with precisely what you wanted out of your immediate?

On occasion, you’ll strike upon the proper immediate, one which delivers a surprisingly thorough, incisive, or in any other case efficient reply. The trick is studying find out how to repeat that success.

Step one: Cease. Don’t simply transfer on to a follow-up immediate. As an alternative, take a beat and ask ChatGPT (or whichever AI chatbot you occur to be utilizing) why your immediate labored so properly.

The subsequent time you get an ideal reply from an AI, do this immediate instantly afterward:

“That was precisely what I wanted. Two issues. First, clarify what about my immediate made this work so properly. Second, flip my request right into a reusable, fill-in-the-blank template I can use subsequent time and inform me which components matter most.”

One factor that units this immediate aside from others I’ve lined is its simplicity. There’s no coding tags, roleplaying, or “be a panel of consultants” stuff right here. As an alternative, it’s an easy “get the recipe” immediate.

Additionally, this immediate is asking for 2 issues that truly work hand-in-hand: a proof for why the immediate labored and a template for doing it once more.

That “create a template” is doing double-duty right here, supplying you with each a blueprint for repeating the immediate’s success and guaranteeing the AI offers you concrete particulars, somewhat than merely flattering you in your prompting prowess.

Right here’s an instance of a ChatGPT immediate (utilizing GPT-5.5 Prompt) that yielded a spot-on response:

“Assist me write an e-mail asking my landlord to repair the dishwasher. We’ve got a pleasant, however barely tense relationship. He’s gradual to reply and I don’t need to come off as demanding, however that is the third time I’ve requested. Hold it brief and well mannered, however with a transparent deadline.”

Once I adopted up with the “clarify what went proper” immediate, right here’s a snippet of the reply I acquired:

What made your immediate work so properly is that it specified the decision-making constraints, not simply the duty.

Most writing requests appear like this:

“Write an e-mail asking my landlord to repair the dishwasher.”

That’s sufficient to provide one thing grammatically appropriate, nevertheless it leaves quite a lot of vital selections as much as the AI.

Your immediate stuffed within the lacking context:

Viewers: “my landlord”
Relationship: “pleasant however barely tense”
Behavioral context: “He’s gradual to reply.”
Your aim: “I don’t need to come off as demanding.”
Related historical past: “that is the third time I’ve requested.”
Output constraints: “Hold it brief and well mannered however with a transparent deadline.”

Every of these eradicated an ambiguity.

Then ChatGPT generated the reusable template:

Write a [type of message] to [recipient].

Relationship: [How we normally interact.]

Context: [What happened and any relevant history.]

Purpose: [What I want them to do.]

Issues to steadiness: [Examples: friendly but firm, apologetic but confident, professional but warm.]

Constraints: [Length, deadline, tone, anything that must or must not be included.]

Mixed, the “why did it work” rationalization and “make me a template” request reveal the successful technique behind the immediate: it focuses on the “what” somewhat than the “how.”

Do that “get the recipe” immediate with different prompts that labored properly for you, and hopefully you’ll uncover new pearls of AI prompting knowledge.