Power Fx - On or Off?

(0) Share

Report

Posted on by Woody Chandler

Just wondering as a question, do people prefer to have Power FX turned On or not? Does it have that big of an impact either way?

Categories:

Power Automate Desktop

I have the same question (0)

All responses (5)

Answers (0)

Suggested answer

eetuRobo 4,484 Super User 2026 Season 1 on at

Like (2)

Report
Copy link

Link copied!

Off

Power Fx was introduced later and seems to still be lacking in some areas especially data table handling. The syntax is very different in power fx vs non power fx regular flows.
I have still yet to find any reason to use Power Fx enabled flows ( if someone has one in mind please share). Only reason I can think of is that if you have done a lot with Power Fx in Power Apps or Excel then maybe as a beginner it might be easier to understand. Or maybe if you have some flow where you would otherwise use some power fx functions in Excel cells and with Power Fx enabled flows you could just do that inside the flow. But I think PADs power fx does not support all power fx functionalities that are in excel so even then you might not be able to do the flow without excel.

Most blog posts, youtube tutorials, documentation and posts on this forum will use regular non power fx flows. So its gonna be easier to learn Power Automate Desktop by using the regular flows.

1 people found this reply helpful.

Was this reply helpful? Yes No
Suggested answer

Pstork1 69,125 Most Valuable Professional on at

Like (1)

Report
Copy link

Link copied!

In general I prefer to have it turned On. When working with variable values Power Automate Desktop has almost no support for functions. PowerFX adds functions that let you process values stored in variables more quickly. For example, there was a post the other day about how to separate out a first name and last name from a single fullname variable. Without PowerFX that requires an intricate process of evaluating each character in the string to identify all the spaces. With PowerFX its as simple as using the Split() function to separate the names into an array and then Last() or First() functions to get the first and last name.

PowerFX is the function language for Power Apps and will eventually be brought into cloud flows as well. I would recommend turning it on unless you have a specific reason not to.

----------------------------------------------------------------------------------
If this Post helped you, please click "Does this answer your question" and give it a like to help others in the community find the answer too!

Paul Papanek Stork, MVP
Blog: https://www.dontpapanic.com/blog

1 people found this reply helpful.

Was this reply helpful? Yes No
Suggested answer

eetuRobo 4,484 Super User 2026 Season 1 on at

Like (2)

Report
Copy link

Link copied!

Pstork1

For example, there was a post the other day about how to separate out a first name and last name from a single fullname variable. Without PowerFX that requires an intricate process of evaluating each character in the string to identify all the spaces.

I don't think it requires very intricate process to do this in regular flow. You just use "Split text" -action like in Power FX and then take the first and second value from the list. Like so:

Sure you can argue that with Power FX you can do that 4 action flow in 3 actions since you don't need to use the Split in its own action but combine it with the First and Last -functions.

Biggest reason I would still recommend using regular flow is that datatables are not as functional in Power Fx since they get very messy if you try to insert data into them. In this example first row was added in the "Create new data table" -action which looks normal but then the "Insert row into data table" inserts a custom object into the rows columns which makes the table not useable.

It can be a preference thing and maybe in some cases its actually wise to use Power Fx like if you need to do a lot of / loop logic like the Fullname splitting then the flow might be more efficient since you can do somethings with less actions.

1 people found this reply helpful.

Was this reply helpful? Yes No
Pstork1 69,125 Most Valuable Professional on at

Like (1)

Report
Copy link

Link copied!

I would suggest using 4 steps instead of Power FX and one or two steps makes it unnecessarily complex. Also, your solution won't work if the fullname contains a middle name. Using First() and Last() works without having to check the size of the array produced by the Split.

For your datatable example the simple fix is that Variables created by PowerFX are objects, not simple strings. Referencing .value on the values you are inserting should solve the problem.

But the bottom line is that PowerFX is where MS is going with functions for the Power Platform. I would suggest getting to know them now rather than later.

----------------------------------------------------------------------------------
If this Post helped you, please click "Does this answer your question" and give it a like to help others in the community find the answer too!

Paul Papanek Stork, MVP
Blog: https://www.dontpapanic.com/blog

1 people found this reply helpful.

Was this reply helpful? Yes No
cwischnewski 160 on at

Like (1)

Report
Copy link

Link copied!

In my opinion I think the answer depends on how much work converting existing bots you have to do. In my case I have a rather large library of bots that I would need to convert over to PowerFX. The alternative is to maintain bots where some are PowerFX and some are legacy, the mental shift needed by the team each time that an existing bot needed attention and then back for new bots doesn't seem worth it.

It is a valid point to assume MS will eventually decide that PowerFX is the only way to create new bots. But without some tool to convert the older inventory we have, no thank you.

As for suggesting that handling of the split list is still a problem when you have variable number of name elements, not true. Once you use split and have a list, the list includes a property for the number of elements. So it should be something like SplitNameList[0] and SplitNameList[SplitNameList.count - 1]. This just illustrates that there are minor differences between the two approaches but most things can be done with both. Also, if you do run into anything that seems impossible with either approach, you can always employ a script.

Was this reply helpful? Yes No