Building Your First Prompt Library

AI is a conversation, not a vending machine.

What you'll leave with

By the end of this lesson, you'll know how to iterate on AI responses through follow-up prompts — rather than giving up or starting over every time the first answer misses.

Why this matters

The first answer AI gives you is rarely the final one. This surprises people who are used to search engines, where you either find what you need or you do not.

With AI, the answer is a starting point. The conversation is the work.

The idea

When the first answer misses, the instinct is often to rephrase the original prompt and try again from scratch. That is sometimes the right move, but there is a faster approach: tell AI exactly what was wrong with the last answer and ask it to fix that specific thing.

This is called iterating. Instead of starting over, you build on what you have.

The teaching block

Useful iteration prompts:

• "That is too long. Cut it in half and keep the most important points."
• "That is too formal. Make it sound more like something a real person would say."
• "You missed the main point. The key thing I need to communicate is [X]. Try again with that at the centre."
• "That is close but the tone is wrong. I need it to sound [warmer / more direct / less corporate]."
• "Give me three different versions of that — one short, one medium, one more detailed."
• "That is a good start. Now add a sentence at the end that [specific addition]."

You can also ask AI to reflect on its own answer:

• "What did you leave out of that summary?"
• "What are the weaknesses in that plan?"
• "What would make this email better?"

These prompts help you use AI as a thinking partner, not just a text generator. You are not accepting the output. You are working with it.

One practical rule: if you have iterated three times and the answer still is not what you need, step back and rewrite your original prompt with more detail. Sometimes the problem is not in the iteration — it is in the original ask.

Example

Show a real iteration sequence. Start with a mediocre first output. Apply two or three targeted follow-up prompts. Show the final result alongside the first one.

The arc: weak start, specific feedback, improved output, one more refinement, solid result.

Try this now

Take any output AI has given you — in this module or before — and deliberately find one thing you would change. Write a single follow-up prompt that asks for exactly that change.

Notice how specific your feedback has to be to get a specific improvement. That specificity is the skill.

Save this

Quiet takeaway

Module 2 is complete. You have the formula, the task-type setups, and the iteration habit. The conversation is the work — and now you know how to have it.

Next

Module 2 is complete. You have the formula, the task-type setups, and the iteration habit. In Module 3, all of that goes to work on the part of your life that might benefit most: everyday tasks at home.