> For the complete documentation index, see [llms.txt](https://examind.gitbook.io/v1/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://examind.gitbook.io/v1/our-approach/why-iterative-feedback-works.md).

# Why Iterative Feedback Works

Skill grows through **practice, feedback, and iteration**. Learning research consistently points the same way: people improve fastest when they attempt work, get specific feedback on it, and revise — over and over.

The bottleneck has always been the feedback itself. Detailed, specific feedback is slow and labor-intensive to produce, so students rarely get more than one round per assignment — and usually only after it's graded, when there's no longer a reason to revise.

[Feedback Machines](/v1/feedback-machines/get-started.md) removes that bottleneck by making detailed, criteria-aligned feedback available **instantly and on demand**. A student can run many practice → feedback → revision loops on a single piece of work, and more loops generally means more growth — whether the skill is persuasive writing, quantitative analysis, or visual communication in a presentation.

That's why Feedback Machines is built around iteration rather than a single verdict: the goal isn't only to grade work, but to help students get better at it.