If you use ChatGPT for the same kind of task again and again, you've probably noticed the same small problems show up each time. Maybe the writing feels inconsistent, an editing workflow keeps missing obvious improvements, or your image generations never quite match from one result to the next. Instead of fixing those problems manually every time, you can improve the workflow itself.
This guide shows you how to do that using the Loop Method, a simple process for reviewing a ChatGPT workflow, finding its biggest weakness, making a focused improvement, and testing the result again. We originally used this approach to improve our video editing and image generation workflows. You can still apply it to writing, design, research, automation, or almost any repeated task you run with ChatGPT.
Tools you'd need to get started:
You only need a couple of things before you begin:
The ChatGPT desktop app with Work or Codex, plus access to sub-agent delegation in your workspace.
A skill, prompt, project folder, or repeatable workflow that already gives you some kind of result.
By the end of this guide, you'll learn how to:
Identify a flow that needs improvement
Instruct ChatGPT to run three review loops
Give it a clear goal
Run the loops and review the report.
Let's get right into it!