Skip to content
Stop starting every ChatGPT conversation from scratch — this one habit saves me time every week
TechRadar
TechnologyTechRadar··3 min read

Stop starting every ChatGPT conversation from scratch — this one habit saves me time every week

Most people use ChatGPT like they’re walking up to a stranger and starting a new conversation every single time.

You open a new chat, explain what you need, add a bit of context, correct the tone, ask it to be more concise, then finally get something useful. And then, the next time you need the same kind of help, you do the whole thing all over again.

That’s fine if you’re asking something random each time. But if you use AI regularly for the same kinds of tasks, it’s a surprisingly inefficient way to work.

The solution is simple: stop starting from scratch.

Instead of treating every chat as a blank page, you can build reusable conversations that already know the job, tone, and desired output. Here’s how I use that one habit to save time every week.

A simple trick

So, we know what we want to do — remove repeated setups for prompts, make the AI’s first answer better, and reduce the amount of refining you need to do afterward. But how do we do it? Here’s the clever bit - you get the AI to do it for you.

Load up one of your previous chats that was productive and worthwhile. Then at the end I want you to copy and paste this:

Sign up for breaking news, reviews, opinion, top tech deals, and more.

“Turn this conversation into a reusable prompt I can paste into a new chat next time. Include the goal, tone, format, constraints and the steps you followed. Make it general enough that I can reuse it with similar tasks.”

Your AI will give you a handy prompt you can use whenever you like now to get a chat that’s exactly what you want.

My meal planner prompt

I now have a reusable prompt for meal planning. Instead of starting from scratch every week, I paste in a short prompt that already knows the rules: I want quick family meals, nothing too expensive, leftovers if possible, and a shopping list organized by supermarket section. Then I add whatever’s in the fridge and how many nights I need to cover.

Here it is:

“You are helping me plan practical family meals for the week. Prioritize meals that are quick, affordable, not too fussy, and likely to produce leftovers. Ask me what ingredients I already have before suggesting anything. Then give me:

A simple meal plan

A shopping list grouped by supermarket section

Any ingredients I can reuse across multiple meals

One backup meal in case I don’t feel like cooking”

This has saved me so much time planning meals in the evening.

A tool that remembers you

Before you get too carried away with this idea, remember that reusable prompts are useful but not infallible. A reusable prompt is only worth saving if it produces better work, not just more predictable work. If your saved prompt is too vague, too bossy, or based on a weak workflow, you’ll just get the same mediocre output, but delivered faster.

Having said that, using reusable prompts has been the biggest change I’ve made to how I use ChatGPT. When a conversation works, I don’t let it disappear. I turn it into a reusable starting point.

If you stop starting every ChatGPT conversation from scratch, ChatGPT starts feeling much less like a conversation with a stranger and more like a chat with a tool that actually remembers how you work.

Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds.

Graham is the Senior Editor for AI at TechRadar. With over 25 years of experience in both online and print journalism, Graham has worked for various market-leading tech brands including Computeractive, PC Pro, iMore, MacFormat, Mac|Life, Maximum PC, and more. He specializes in reporting on everything to do with AI and has appeared on BBC TV shows like BBC One Breakfast and on Radio 4 commenting on the latest trends in tech. Graham has an honors degree in Computer Science and spends his spare time podcasting and blogging.

View original source — TechRadar