Your flow's performance may be slow because it's been consuming more data than expected since...

(0) Share

Report

Posted on by enriqueglopez

503 Moderator

Hello everyone, I have the following error:

Your flow's performance may be slow because it's been consuming more data than expected since 5/14/2025 11:00:59 AM (1 week ago). Learn more

My flow is recurring and runs every 30 minutes. It retrieves items from Dataverse and performs an "Apply to each" action where it updates them based on other data.

I have a large database being queried — my flow was intended for a one-time migration to apply metadata to items that were missing it. Initially, there were around 60,000 items, but the idea was that with each run, the number of items would decrease. At some point, the flow couldn’t handle it anymore and has been stuck like this ever since.

This message appeared, and it doesn't go away even if the flow is turned off or paused. When the flow is reactivated with lower recurrence or fewer items, it still doesn't work. It runs for hours or days without doing anything.

How can I fix this?

Thank you

Categories:

Building flows

I have the same question (0)

All responses (7)

Answers (1)

stampcoin 5,058 Super User 2025 Season 2 on at

Like (0)

Report
@enriqueglopez Hi,

Do you have the correct license for the flow ? Since you already checked the link, I assume you have.

Some suggestions:

with correct license, if you still have the problem, consider splitting the task with two account, each of the take 12 hours shift, which balance the API consumption.

on your source , create a flag to indicate the if migrated or not. so the flow only takes the needed ones.

Use static variable if possible, for example each 30 min only take N items.

review your flow if anything can be optimized.

good luck.

Enjoy Power Platform and have a great day 🚀 | My LinkedIn

If the answer helps, please consider 👍, Thanks.

Was this reply helpful? Yes No
enriqueglopez 503 Moderator on at

Like (0)

Report

@stampcoin Hello, I've already implemented the steps 2 and 3, thank you so much for your advice. I'm worried because that is everything I can think of and still not working well...

I don't know if my flow's license is correct. I have the Power Automate Free, and my flow is configured as follows:

This is my flow:

I've set a variable (DataverseResponse) with the dataverse response (big list, 52k items) and compare the items iterated (2000) per flow execution. I thought this would reduce the quantity of actions performed.

Let me know if you have any question or tip. Thank you so much.

Was this reply helpful? Yes No
Chriddle 8,443 Super User 2025 Season 2 on at

Like (0)

Report

You might consider batch operations.

Dataverse Connectors are premium.

Was this reply helpful? Yes No
stampcoin 5,058 Super User 2025 Season 2 on at

Like (0)

Report
@enriqueglopez Hi,

One thing you have to change is the pagination threshold value, it will consume your more API.

for example you want to fetch 2000 each hour, you 2000 as the value for pagination.

current one you are wasting API consumption.

So change it on the top, the whole flow use the same.

the dataverseResponse can be removed if you can narrow down the list size (2000 for example).

Other optimization will be ( kind of re-do task):

use a temp resource outside of the flow, for example.

use data flows.

use other api .

feel free if you have any questions.

Was this reply helpful? Yes No
enriqueglopez 503 Moderator on at

Like (0)

Report

@stampcoin So, is it better to make 2,000 individual calls to the large list using a filter (each returning just 1 item, so 2,000 actions, 2,000 items) instead of retrieving the whole table at once (~56,000 items) and working with the values in a variable?

I’m not sure if I’m explaining it clearly—let me know if it’s unclear.

Was this reply helpful? Yes No
Verified answer

stampcoin 5,058 Super User 2025 Season 2 on at

Like (0)

Report

@enriqueglopez Hi,

Yes—it is better (for both performance and reliability in Power Automate) to make 2,000 individual filtered calls (each returning just 1 item) than to pull the entire table of 56,000 items into a variable and iterate/filter that in-memory.Never fetch all 56k+ records into an array/variable for in-flow comparison.

So, change that one first if you can see the improvement.

Was this reply helpful? Yes No
enriqueglopez 503 Moderator on at

Like (0)

Report

@stampcoin It seems to work better now. Thank you so much!

Was this reply helpful? Yes No