If you are a flow maker then you must become very familiar with the request limits and configuration limits for the product. If you aren’t then you are at risk of your flowing failing or getting throttled.
The two key pages for this info are at:
In the last few days there have been some critical updates to the Request limits and allocations page to clearly explain which account is ‘charged’ for the API calls.
It has been well misunderstood and a source of confusion for many regarding who is charged – the flow owner, the user that invokes the flow or the user account used for the connections.
I want to be totally and absolutely clear for everyone reading this post. The credentials of the user account being used for the action will not be the user account that is charged with the API calls. This is the source of most the confusion.
With the addition of the following FAQ item it should be clear to all (finally) that the API calls are charged as follows.
“It depends if the process is run on-demand or in the background. Instant flows, which are run on-demand, will use the limits of the account who started the process. On the other hand, workflows or automated/scheduled flows that run in the background will always use the limits of the owner of the process irrespective of why the process started or what accounts are used for connections inside of the process.”
And straight from Stephen Siciliano, the Microsoft Power Automate Czar
“If it’s triggered in the background, e.g., on a schedule or based on a cloud event – then it’s from the owner’s limits.
If it’s triggered on-demand, e.g., from a Power App, then its from the person triggering it’s limits.”