Understanding your ROIP score
A score from 0 to 100 that you can trace back to your prompts, your turns and the result.
Your ROIP is a number from 0 to 100. Higher means your prompts gave the AI more of what each task needed, with less left for it to guess.
Where it comes from
- Task type: what kind of work the session is, such as Email & Message, Debug or Research. It decides which components count, and how much.
- Roles: what each prompt does (New Ask, Refinement, Correction, Extension or Close) changes which components it's graded on.
- Turn grades: each prompt's grade is the weighted share of the components it needed that it gave.
- Components: each turn shows which components it had and which it was missing, so you can see where the points went.
What a score doesn't mean
A lower score isn't a judgment of the result you got. It shows how much the AI had to guess, and where putting more in your prompts would have saved it.
A Correction is graded like any other turn, even when the AI made the mistake. One that says what's wrong and how to tell it's fixed scores well; a bare “that's wrong” scores low.