🎁 Get an indication of process (in)efficiency with AI Studio

English Forum AI Studio
1

A feature is a means to an end. It’s never about the feature, but what it enables you to do.

There are a lot of reasons why you’d want to be able to Increment a custom number field with a rule action

The simplest to implement would be Tracking the number of times a task changes between section

The reason for wanting this would be to have it serve as an indicator of process (in)efficiency. If tasks in a process regularly switch sections more often than there are steps in the process this is probably worth investigating.

This is how to set it up:

  • Create a custom number field section movements
  • Create a rule with AI studio like so:

image
image1341×860 42 KB

  • Although GPT-4o is probably be the most accurate model for numerical operations this seems overkill, so I’ve selected GPT-4o mini
  • Set the instructions to section movements = section movements +1 to be as clear and unambiguous as possible
  • Restrict the data and permissions as pulling information from other work/projects is unnecessary and would therefore be wasteful.
  • Set the rule name to 📝 Log section movement. I often use an emoji to easily spot which rule was running or has run:

image
image260×116 2.91 KB

image

:warning: A few caveats :warning:

  • We can never be as sure of the accuracy of this rule as when Asana would implement incrementing a number.
  • As computing power is concerned I can hardly think of a more expensive/inefficient way to increment a number. This is why I’d advise to use this sparingly, and only as long the process you’re evaluating has focus under an improvement initiative.

Jan-Rienk @ Improving Every Day

4 Likes
2

We are still waiting to see the actual quotas, thresholds and pricing, but there is a world where this is a bit too expensive to do compared to the benefits. I hope we get those answers soon!

3

This (and the environment) is a reason I was hesitating to post this.

Still I think this illustrates one many situations where when the required feature doesn’t exist yet, and AI can be used to fill the gap.

Still these trade-offs should be made wisely.

1 Like