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 This announcement was originally made at Inspire this year, but has undergone a number of significant changes.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/

[Pstork1]

I think the answer is somewhere in-between the $40 license and a CDS and D365 connector only.  Its all about the application context.  If the app "extends a Dyanmics application" but uses a non-dynamics premium connector I believe its still covered just like it was in the past.  What won't be covered are standalone applications that have nothing to do with Dynamics.  Its all about the application context.  Here's the section from the FAQ.

Will PowerApps and Microsoft Flow use rights change for Dynamics 365 applications?

PowerApps use rights with Dynamics 365 licenses: Dynamics 365 Enterprise licenses will no longer include general purpose PowerApps capabilities. Dynamics 365 Enterprise users will continue to be able to run apps and portals that extend and customize the licensed Dynamics 365 application, as long as those apps and portals are located in the same environment as their licensed Dynamics 365 application. Custom apps or portals outside of the Dynamics 365 environment will require a standalone PowerApps license.

Microsoft Flow use rights with Dynamics 365 licenses: Dynamics 365 licenses will no longer include general purpose Microsoft Flow capabilities. Microsoft Flows will need to map to licensed Dynamics 365 application context - Microsoft Flows should trigger from OR connect to data sources within use rights of licensed Dynamics 365 application(s). Use of standalone flows will require a Microsoft Flow license.

[Anonymous]

That is exactly what I was looking for, although not the answer I was hoping for!  We do all of our custom canvas apps outside of the context of our D365 environments.  We work with all custom procsses and do no extending of D365.  So it looks like the places where we have used Dynamics licenses to get PowerApps P2 and use premium connectors are all going to be impacted.  That is a whole different issue for us that may not even fall under being grandfathered.  I've definitely got a lot of fun in the next few months, and I hope I still have a job in the few after that!  We built out a department with 4 developers using PowerApps and its usefulness is slowly slipping away.  Oh joy.

[Pstork1]

Existing Apps wherever thay are that use SQL will still be covered by the grandfathering for 5 years. But any new apps would require either App only or standalone user licenses.

[seanbrogan]

If I create 15 blank apps tomorrow and connect them to my SQL database, will they be grandfathered for 5 years too? And I'll still be able to develop them further within that 5 years?

[Pstork1]

I've already addresssed this earlier in this thread.  The answer is that Yes, theoretically that would give you 15 grandfathered apps.  It would also be a clear attempt to violate the spirit of the new licensing rules. So I don't recommend it.

[seanbrogan]

Thank you for your reply.

To be honest I dont really want to do it. But realistically, taking away a service I am already using violates the spirit of the licenses I am already paying for.

[DavesTechTips]

Hello everyone.

I mostly don't enjoy the questions that clients are now asking: "What are the alternatives?" Please please please we don't want to go back there!

The second question from clients that cause for some very uncomfortable conversations: How can we confidently continue with full deployment of Teams, Planner, etc, if we might be faced with the same problem in a little while? Erm erm erm erm, look a squirrel!

I appreciate what Microsoft is doing and the fact that they sometimes need to make changes to the pricing models especially when introducing new functionality like portals and external access. That being said, everyone's worst nightmare is that once everyone is using Office 365 (which they will), the licensing models would become exponentially more expensive as Microsoft has 'locked' everyone in. I don't want to believe that this is the case it doesn't seem coherent with every talk I see form Satya and the great work that Microsoft is doing in the community. This is however clearly affecting clients' willingness to just implement O365 services as it makes perfect sense.

To me this conversation can be split in two. The way the licensing model approaches new functionality and the way it affects existing functionality. The new functionality is fair game as no one had this in the past, so I am just talking about existing functionality for now.

Concern 1) The apparent frequent changes in licensing is making clients extremely uncomfortable for long term investment. I think this is having a bigger impact that anyone anticipated.

Concern 2) Users that might need access to one or two premium connectors (or SQL Azure now), but nothing else are now in for some significant costs, even if they don't need any portals or Model-Driven functionality. If someone for an example uses the Word online connector, they would have been able to get a P1 in the past, but now they have to opt in for the per user per app per month license which could potentially be just as expensive as the P2 or per user license if they are using a few apps. There seems to be a big gap on this one as the only way they would be able to get access to that one premium connector (or SQL in Azure now), would be through the per user per app per month model that also gives access to model driven apps, portals, external access and sushi! It feels like this shouldn't be the only option to get access to a premium connector. It might be more cost effective for this user to then use the pay per action model from Logic Apps, but they might want to access this directly from PowerApps depending on the connector.

Concern 3) Another concern, very similar to the previous one, is that sometimes we use HTTP actions to enrich the functionality between existing Office 365 workloads. An example of this is making use of some more advanced filter queries in Flow, and then dynamically return the data to PowerApps from Flow using an HTTP response as per one of ours video on http://youtube/davestechtips. This would effectively mean that this user would have to buy a per user per app per month license to make use of this functionality.

We need a more cost effective and manageable way to get access to premium connectors and on premises data sources directly from Canvas apps (and SQL Azure). Originally the HTTP and custom connectors were accessible to the Office 365 entitlement (limited to one per user) but this changed in March 2019. Maybe bring this back where a normal user can access one or two premium functions as part of the O365 entitlement. Maybe also consider a similar model to Logic Apps, allowing per action billing and another one which just allows access to premium connectors, without all the other goodies from portals, models and sushi. 😉

The PowerPlatform has an opportunity to be the most connected system in the world and if we can get a workable and cost viable solution for various connectors, I don't see a reason why this opportunity has to be shot down. I have been telling people that PowerApps will be to business applications, like what PowerPoint is to presentations. I hope I wasn't wrong.

Dawid van Heerden
Follow on Twitter: @davestechtips
Subscribe to YouTube: https://www.youtube.com/davestechtips?sub_confirmation=1

[iAm_ManCat]

I will be honest here. I have created only Apps that use SharePoint as a backend, however that obviously has its limits. The goal was to start using AzureSQL at a later stage with complex business Apps. There are/were plans for an additional 50+ Apps that would replace business processes (complex ones) that would need a proper relational database and would need to not be limited in terms of number of rows (SharePoint I'm looking at you).

I tried working out what it would cost since these Apps will be shared with 600+ staff. On either side of the licensing model (User or App) there is not a chance that I can justify the additional licensing expenses on that to my bosses.

I have now started teaching myself C# again, and will instead ask them for money for courses, as this will be the most cost-effective route for in-house development. At least I can hook into our existing data sources using API's etc without having to pay to make that connection. Plus I can then just use the Graph API to poke anything on the M365 platform, again without additional costs (or at least very minimal costs).

This change makes me really sad, as I had planned to continue my career path down the PowerApps route.

..but I can hear the discussions going on about alternatives already as the licensing costs would end up being an additional 30% of the entire IT budget for the last year - could anyone justify the value being gained as equating to that?

By comparison, we could hire a team of 4 full-time devs and have money to spare for infrastructure costs and training. Management may decide that my role no longer gains them benefits if the licensing cost prevents the plans for future developments along this route (my main concern).

I'll still keep using the platform for seeded-level development (as long as that's not separated out for licensing eventually?) -standalones with sharepoint backend if they aren't expected to generate lots of data - but if I want my career to progress I'm going to have to go back to learning traditional App Dev.

[JohnP]

---

@Pstork1 wrote:

I've already addresssed this earlier in this thread.  The answer is that Yes, theoretically that would give you 15 grandfathered apps.  It would also be a clear attempt to violate the spirit of the new licensing rules. So I don't recommend it.

---

Spirit? You think MS has a spirit? Or a soul? They are a greedy multi-billion corporation. I would recommend this practice all day. I'm currently exploring creating an SQL elastic pool with a 100 (or more) idle databases. Then create 100 (or more) empty apps that connects to all those databases. Then I'll be grandfathered for 5 years until I can think of a better alternative.

[Anonymous]

@iAm_ManCat  This is one of the best posts I have seen on this whole thread and I think it really captures where I am at as well, our business is a little afraid of getting involved in dev ops, and PowerApps was helping massively with that, but now it looks like to get anything real done we will have to rely on suppliers and pay and wait, or push the issue and see if we can get an in-house dev ops team using c# and what not. 

This was a major gap that powerapps totally plugged, and they have just priced themselves out of the gap - with the amount of disatisfied people from this change, it makes it clear that this is the case. 

What would be really cool to see is if all the people who have been let down got together and made their own low code platform (Not proposing, just saying it would be cool and almost unsuprising to see)