Introduction

Is your Power Automate flow taking too long to run? Slow flows not only increase execution time but also consume more API calls, making automation less efficient.

In this article, we'll explore 10 common mistakes that slow down Power Automate flows and learn the best practices to improve performance.

1. Nested Apply to Each Loops

Problem

Using an Apply to each loop inside another loop increases execution time dramatically.

Example

Loop through Orders
Inside each Order, loop through Order Items

Better Approach

Use Filter Array
Use Select
Restructure your data before looping

Benefit

Faster execution
Fewer iterations

2. Retrieving Everything Then Filtering

Problem

Many developers use Get Items to retrieve every SharePoint item and then apply Filter Array.

Get Items
↓
Filter Array

This downloads unnecessary records.

Better Approach

Use an OData Filter Query

Example:

Status eq 'Approved'

Instead of retrieving 5,000 items, retrieve only the required records.

Benefit

Lower API usage
Faster execution
Less memory consumption

3. Multiple Get Item Actions

Problem

Calling Get Item repeatedly for the same record.

Example

Get Item
Get Item
Get Item

Better Approach

Reuse outputs from previous actions using dynamic content or Compose.

Benefit

Fewer SharePoint calls
Reduced latency

4. Overusing Variables

Problem

Using variables inside loops.

Set Variable
Set Variable
Set Variable

Variables increase processing overhead.

Better Approach

Use

Compose
Select
Filter Array
Expressions

instead of variables whenever possible.

Benefit

Cleaner and faster flows.

5. Sequential Loops

Problem

Processing records one by one.

Example

500 records
1
2
3
...
500

Better Approach

Enable Concurrency Control.

Multiple records can be processed simultaneously.

Benefit

Significant reduction in runtime
Better throughput

6. Triggering Too Often

Problem

The flow starts every time an item changes and then checks a condition.

Example

Trigger
↓
Condition
↓
Terminate

Most runs do nothing.

Better Approach

Use Trigger Conditions so the flow starts only when required.

Benefit

Saves API calls
Reduces unnecessary flow runs
Improves performance

7. Large Attachments

Problem

Downloading large attachments repeatedly.

Better Approach

Filter files first
Download only required files
Avoid unnecessary attachment processing

Benefit

Less download time and reduced execution duration.

8. Excessive Connector Calls

Problem

Repeatedly calling connectors like:

Get User Profile
Get Manager
Office 365 Users

inside loops.

Better Approach

Retrieve data once and reuse it throughout the flow.

Benefit

Lower API usage
Reduced throttling
Faster execution

9. Connector Calls Inside Loops

Problem

Calling SharePoint or Dataverse actions inside every iteration.

Example

Apply to Each
Get Item

If the loop has 1,000 items, that's 1,000 API calls.

Better Approach

Retrieve data once before entering the loop and reference it using expressions.

Benefit

Massive reduction in API calls
Better scalability

10. No Error Handling

Problem

If one action fails, the entire flow stops.

Better Approach

Use

Try
Catch
Finally

with Configure Run After.

Benefit

More reliable flows
Better troubleshooting
Prevents downstream failures

Bonus Performance Tips

✔ Use OData Filters instead of Filter Array when possible.

✔ Enable Concurrency Control for large loops.

✔ Retrieve only required columns using Select Query.

✔ Avoid unnecessary variables.

✔ Minimize connector calls.

✔ Reuse outputs instead of calling actions again.

✔ Use Trigger Conditions to prevent unnecessary runs.

✔ Add proper error handling.

Conclusion

Most Power Automate performance issues are caused by design choices rather than platform limitations. By reducing unnecessary loops, filtering data at the source, minimizing connector calls, enabling concurrency, and implementing proper error handling, you can build flows that are faster, more reliable, and more scalable.

Following these best practices will help reduce execution time, lower API consumption, and improve the overall performance of your Power Automate solutions.