New PowerApps and Flow Licensing coming October 1, 2019

[Pstork1]

If you haven't seen the announcement yet about the Licensing changes that are coming to PowerApps and Flow starting on October 1, 2019 you really need to read the following Blog. This announcement was originally made at Inspire this year, but has undergone a number of significant changes.  Make sure you read this and understand what it means to you and your organization.

https://powerapps.microsoft.com/en-us/blog/new-licensing-options-for-powerapps-and-flow/

[WLAS_Almighty]

From someone that have checked and tested almost everything else that is out there...

In the Business Application Development paradigm:

Outsystems, AwareIM, Mendix, Caspio, Kintone, AppSheet, QuickBase, Kony, Wavemaker, KissFlow, Zoho Creator, (I can go on and on)...

And in the mobile app development sector:

Appery.io, Roadie, AppBuilder, GoodBarber, AppyPie, BiznessApps....

I can tell you that Microsoft is not stupid. They did not come up with that price out of a whim and just to anger us. they are still VERY competitive (actually cheaper than most - NOTE: if you stay at $10.00/user/month) when it comes to the competition and they had a strong start.

I, myself felt that they had it just right and joined the bandwagon.

But, before giving you my humble advice, let me explain where I am coming from:

I am an ISV looking to mobilize my 20+ year "heavy" MS-SQL database program developed in C#.NET.

I was hoping - since all my clients are Windows based and use Office - to have PowerApps take small, but important blocks of usability and make it prettier and more usable.

But after developing for a while, and with the new 2-app limitation, I see that Microsoft cornered me into a un-winnable situation.

Currently, PowerApps development requires simple screen and logic design. If you have too much going on (with screen popups, validations, information, etc...) it becomes unyielding, cumbersome, slow, buggy.

So, it is not a heavy application design.

But it is not a strictly mobile app either. 

So, I think where they went wrong is to think that people can do what they have been doing within the 2-app limit. If they removed that limitation, they would be better/cheaper than competition.

But to think my users will have to pay $40.00/month to solve basic problems... no. That makes PowerApps a much worse choice than all competitors named above.

So, as a business application developer in need of  on-premise, mobile, reporting, RAD tool (to save on paying thousands on a developer that does not quite get what you want), ability to do anything 9including offline and synchronization), PaaS and SaaS, I chose to go with FileMaker Pro.

I know, I know, I never thought I would say this. They are Apple for Christ sake!

But, for me, they fit the "current" bill perfectly. And although expensive, will be less and much better than PowerApps.

[sscarcella]

I have gone back to using BCS.  It is a bit slower but at least I can access my Azure SQL and SQL Server data in my PowerApps now.  Does anyone know how long BCS will be supported?

[sscarcella]

I have gone back to using BCS just like I did with InfoPath.  It is a bit slower but at least I can access my Azure SQL and SQL Server data in my PowerApps now.  Does anyone know how long BCS will be supported?  We certainly cannot afford any of the 3rd party products that charge $30 -$40 per user per month or the new PowerApps $40 plan.  We have used SQL Server for over 12 years and it seems outrageous that you pay $$$ for Server and then have to pay $$$$ to access the data using PowerApps.  All our business processes and forms use data from SQL so the $10 plan is not a solution.  Once they cut off BCS access, we will have to invest in the Layer2 Cloud Connector and sync our SQL tables with SharePoint online.  But they need to improve SharePoint list integration with PowerApps before that is a solution for us.

[tarinalees]

We chose FileMaker as well. @WLAS_Almighty , thanks for your addition to the post, seeing it spelled out by you, and why you also chose FileMaker, makes me feel really good about our choice. Good luck everyone! 

[WillPage]

I have a question about the API limits in the new licensing construct. The documentation states that they consider an API call to be (amongst other things) a Flow action. For connections like SharePoint or Exchange Online I can understand that, but what about data operations and variables?

What if I have an array of 1000 items and I want to loop through each item and perform a condition within the loop that appends to a string variable or an array variable depending on the value of some attribute within each object in the array. I'm not hitting any APIs doing that, I'm just building a string or an array based on data I already hold.

Would each condition and subsequent action within the Yes or No branches of the condition count as an API call each? I would hit the 2000 limit pretty quickly like that.

[Pstork1]

We are still waiting to get some usage reports that can be used to track those kinds of things.  They are coming.  In the meantime, two things.

1) MS says that the new API limits would affect less than 5% of current customers.  So most users will never have an issue if they maintain current levels

2) Batch actions are treated as a single API. Can't tell you exactly what they mean by Batch but that was the point made.

Hopefully there will be more tools and clarity of definition coming soon.

[PaulD1]

---

@Pstork1 wrote:

We are still waiting to get some usage reports that can be used to track those kinds of things.  They are coming.  In the meantime, two things.

1) MS says that the new API limits would affect less than 5% of current customers.  So most users will never have an issue if they maintain current levels

2) Batch actions are treated as a single API. Can't tell you exactly what they mean by Batch but that was the point made.

Hopefully there will be more tools and clarity of definition coming soon.

---

Thanks of the info @Pstork1. 

I think I will wait for the tools and clarity before dismissing my fears over the new API limits. I suspect the majority of current customers are still evaluating and experimenting with Flow or even not/barely using Flow at all, so while only 5% of current customers are affected that could mean quite a high % of current customers who are actively/seriously using Flow will be impacted.

[monkeyclass]

The new licensing model has also really put a stop to current projects developed in our +100 person department, in a +40.000 person company where we where a pilot project. Very unfortunate.

[pulsebeat]

What?  Your company is not willing to pay nearly $20,000,000/year so that people can use "citizen developer" apps to do things like, you know, sign up for courses and track whose turn it is to bring in the Friday donuts? What's wrong with your leadership!   

[JohnP]

The new pricing model does not scale well. Organizations may have thousands of *potential* or *occasional* users for any given app. Forcing licenses on all of them is not fair. The total silence from MS is not very surprising, but still disturbing.

[WillEdz]

This is starting to feel a bit like the moment when Microsoft sent out the message that Access Web Apps was being shut down. That caused a major wobble. The answer from Microsoft was rebuild all your stuff in PowerApps its great and free.... here we are a relatively short while later and the message is "Thanks for rebuilding everything (at your cost) - but since you have its time to pay again - but this time we are really going to make you pay - and we mean pay allot each month"

The questions and raised eyebrows that are being fired towards me are not good. 

Label1.Text="Your"

Label2.Text="Fired"

Text(Label1.Text, " ", Label2.Text)

[pulsebeat]

I prefer the string concatenation operator.

Set(varFirstPart, "I'm screwed.  Good time to");

Set(varSecondPart, "retire")

varFirstPart & " " & varSecondPart

[FredericForest]

Considering the utter lack of response from MS on some very genuine concern from the community (not even a "sorry guys but that's that") I'd say there's still some issue with the Communication connector 😉

Notify("Hello world ?", NotificationType.Warning);

[WillEdz]

Careful you might need to upgrade to a $40 a month per user if you use variables 😏

[robinlotz]

Office 365 E5 (Skype (with Phone call );Teams;Power BI Pro;Word;
Excel;Sharepoint;Outlook;OneDrive;PowerPoint;OneNote...) = 35$/user/month

Power Apps "Unlimited" (with ur own dev team / application maintenance team + licensing for ur cloud data base on AZURE ofc ) = 40$/user/month

Just saying...
The pricing before the update of the premium SQL connector was ok ...

[Woods]

Consider myself lucky that I didn´t go all in down this path. Been looking at it for a year and used a fair amount of hours learning the basics in both BI/PA in my free time. I was so engaged. "This is really the future" i told my partners. Now I feel really sorry for all the consultants out there selling solutions based on the Power platform. You're product is now sh*t because of price. I feel really sorry for all the people used countless of hours creating awesome in-house solutions that now will be non-sustainable. You're product is now sh*t because of price. Both of you will have serious problems explaining this for your customer/ management.

What the heck MS?! You know you have to work hard to gain trust after this blunder? Even if you turned down prices cheaper than before, i wouldn't touch it without a guarantee on 5-10 year pricing for my company. For a long time after the new price-model was out I thought there was something i had missed, because who want to pay that amount of money? I run a consultancy company, and have been working with developers for the past 5 years. They are expensive, and too often act like they're God himself, but they at least got integrity and for the most part i can trust them. 

All those hours down the drain... Still I consider myself lucky, because I bet It's nothing compared to most of you. Now what? I still feel this is the future but MS is out until they've cleared this mess! What is the best solution compared to Power Apps?

[FredericForest]

Just saw the MS Ignite presentation on Power Apps licensing. Boy that made me sad... 

Once there was beautiful platform full of potential. Then one day, the Licensing fairy came and cast a wicked spell on her...

The end.

[strategery]

I learned of this via pop-up when a user was trying to start an app. So my questions are

1) Is it apps created before Oct 1 that get the grace period or published before Oct 1?

2) Can we safely ignore this pop-up or is it indicative the app will be rendered useless soon?

Just my 2 cents.

I've been a contributor on the forums and helped others, spent many hours developing my own skills on this product. We already pay an Office365 subscription. When "on-premises" sources were deemed premium, I thought "well, it kinda makes sense...payback for developing the gateway, etc.". I convinced them to get Azure SQL by subscription, made sense, an easy sell.  Now this feels like the rug has been pulled from under us. So we pay for Office365, we pay for Azure SQL, and you want to charge more to bridge the gap? Is Microsoft trying to kill this product?  I'll be moving away from PowerApps if nothing changes, simply for the fact I can't provide solutions without the worry somewhere down the road taxes will be added to it. I'm a developer, not a salesman.

[Pstork1]

The one year grace period should apply to the tenant in general.  But the 5 year extension for SQL is only for apps that existed prior to Oct. 1.

[PaulD1]

Time for a mini-rant.

Earlier this week I have spent considerable time and effort porting an App from D365 to CDS connectors (prompted by the banner in the designer) only to discover there is a known bug with the CDS connector with no resolution other than to keep deleting and recreating galleries until you get lucky and it works (for a while, then you have to go through the process again).

Yesterday in trials of an App, saving an image to Azure Blob Storage from PowerApps takes a couple of seconds on Web and iOS and a couple of minutes on Android - still investigating but very much seems like a platform issue.

This morning the behaviour of Explicit Column Selection has been changed breaking existing production apps used by hundreds of our users.

MS thinks this platform is mature enough to charge $10 per App per User per Month?

[CurtisJohnson81]

@PaulD1  well put, I totally agree that its not ready to be charging at this level and stand up to the competition. 

I could build a case for this in our business but in the back of my mind I am worried about the bugs that I am aware of that I know would hold us back - they did this a year too early I think. One can only hope that any new funds that they get as a result of the licence change quickly results in massive input into the platform which will resolve these issues fast. I hope that MS do this becuase I love the Powerplatform and I want to see it help businesses reach their potential! 

[pulsebeat]

@CurtisJohnson81  

"I could build a case for this in our business"

I would really like to know what that is. I just don't see it, no matter how many ways I work the numbers, no matter what color tinted glasses I put on, no matter how much I squint my eyes.  Maybe I have missed something - some magic coefficient that makes the case.  If you know of one, please do tell.  Like you, I really love the Power platform but at .5 million, I'm now back on the market looking for love in a new relationship.  

[DirectorMumbles]

@pulsebeat only thing I can see at this point is small one off apps. Nobody would use this at any scale anymore.  If I need an app for one or two people, not a big deal. Scale that out to 100 and I might as well not use the platform at all because the ROI is no longer there.

[CurtisJohnson81]

@pulsebeat  hey man I am with you and totally agree, I just got skills in my case building is all - I jest, I jest.

In the specific scenario it does stack up against suppliers, when you are at scale, nothing is cheap, apart from cheap crap. what I agree with is that Powerapps is not the best platform you can get for you money, so building a case, yes, going for it in production - maybe not so much.

besides, if case building was that difficult, marketing wouldn't exist : ) 

[PaulD1]

Time for another round of grumbles (not a good week for me! 🙄)

Client wanted a minor modification to an existing Flow (that has run fine for months) that uses the v1 Outlook connector. Made the change, Flow reported it was running, I could see the emails being sent. However, turns out the emails were no longer being delivered to external (to the org) email addresses! Not even sure how that is possible but I have the evidence of dozens of emails if anyone wants to challenge this.

Reverted to a saved copy from before making the changes, but this had the same problem! Only option, change the Flow to use the V2 Outlook connector, but that does not support the 'Send As' functionality, so now all the emails are coming from the service account until we can put a work-around in place.

At the same time as this I was working on another App which uses D365 connector (we tried migrating to CDS connector but it is too buggy). Modified an entity, removed and re-added the connection to that entity in the App but started getting lots of runtime errors. Solution was to remove and re-add every entity connection, even though they had not been modified.

Even at the old price for Power Platform (i.e. most features free with O365) I'm starting to think that the total cost of development, maintenance and downtime (over the App lifecycle) exceeds the competition for any halfway complex app.

[monkeyclass]

So we took the consequence of the new licensing model and replaced our entire backlog of PowerApps projects, and made it into Django projects. We finished the first project within a month and now how it running it production (hosted as an app service on Azure). The project will also work as the template for the coming projects, which should decrease the development time. It took time to get flying but it seems like the right approach considering we can now develop real web-apps rapidly, and do not have to pay licenses to add extra users. Everyone is validated against our Active Directory and everyone is happy. Goodbye PowerApps.

[pulsebeat]

@monkeyclass  Thanks for the post.  This really  gives  me a good place to start with my PowerApps replacement program.  Best of luck to you.