Iterate: the first answer is a draft

Here's the habit that separates people who get great results from AI from those who get frustrated: they iterate. They treat the first answer as a draft and refine it — and that back-and-forth is where the magic actually happens.

Treat AI like a conversation, not a vending machine. A common mistake is to type one prompt, get an okay answer, and either accept it or give up. But AI is interactive — your first result is a starting point you can steer. If it's not quite right, just say what to change:

• "Make it shorter."
• "That's too formal — make it warmer and more casual."
• "Add a part about X."
• "Explain that last point more simply."
• "Give me three more options."
• "Rewrite it focusing on Y instead."

Each round nudges the answer closer to what you want. Two or three rounds of refinement will get you a far better result than any single perfect prompt.

Why iteration beats trying to write the perfect prompt. People imagine expert AI users craft one flawless prompt. In reality, they refine — they get a draft, see what's off, and adjust. It's much easier (and more effective) to steer a draft than to specify everything perfectly up front. So don't stress about the perfect first prompt; get a reasonable one, then shape the result.

A couple of iteration power-moves:

• Ask the AI to improve its own work: "What's weak about this? Make it better."
• Ask it what it needs: "What would help you do this better?" or "Ask me any questions you need before you start." — great for complex tasks, so it gathers what it's missing instead of guessing.
• Don't settle. If an answer is 80% right, that last 20% is usually one quick follow-up away. Ask for it.

The mindset: the habit that gets great AI results is iteration — treat the first answer as a draft and refine it through a few rounds ("shorter," "warmer," "add X," "give me other options"), because steering a draft is far more effective than trying to write one perfect prompt. AI is a conversation, not a vending machine: get a reasonable start, then shape it toward exactly what you want, using moves like asking it to improve its own work or to ask you clarifying questions. Don't settle for a so-so first answer — the great result is usually a follow-up or two away.