Recently, I got a request, or rather a kind ask, if I could give my perspective on how to grow even more in our small, tiny, little part of the world – the Power Platform.
I know, in the big picture, this platform means nothing in a hundred years, but here and now, we live on this platform.
Luka Radulović started the Serbian Power Platform Community and asked for my help. Thanks for asking me! 🙏
Here and now it has been part of my life the last… hm… is that really 17 years?? Yup, it is. I stumbled into this technology area in 2009, and I’m still living in it. Too big words? Probably. Anyway, these 17 years have given me a bit of both experience and, more importantly, perspective to develop in the Power Platform. I hope a few of my tips are useful for anyone 🙂
I’m honored to receive the ask, and it’s clear to me that I should give back to the community whenever I can.
Here comes the five questions and my advice. Do you agree with my thoughts?
Know solutions and don’t reinvent Power Platform
Looking back at your early Dynamics 365 / Power Platform career, what is one mistake or misconception you see new developers often making and what would you do differently today?
My early days were ages ago; I started in Dynamics CRM 4.0 in 2009, so we’ll see what I remember…
I’ve been a consultant-developer since 1998 and have not worked on any Microsoft-related platform until I stumbled into Dynamics, except that we were (of course) using Windows.
We who work on Power Platform usually come from one background: 1) I’m a true rookie, or 2) like me, with a lot of dev experience and knowing more than we actually need today.
- Not really understanding solutions unmanaged vs managed.
We should always deploy with managed solutions, by me since CRM 2011 supports this, and by Microsoft for the last 5-10 years.
BUT – managed does not mean we can’t change it manually! A lot of people say, “Yes, I can fix your problem in 2 seconds,” and do it manually in the PROD environment. It will mess up the next proper deployment, depending on the parameters for the import.
We must 100% understand this. Well, at least 80%, 100% is utopia… - If I’m a badass developer, it’s easy to think the Power Platform is mainly only a database. Then we can build on what we already know, developing our own approach to UI and ignoring Model-Driven Apps, Canvas Apps, Power Pages, etc. Then we are paying crazy too much, and using a fraction of what the platform can offer.
Today I always start to look at what is possible to solve the issue OOTB, and only if it is impossible or there are too many workarounds that make maintainability a nightmare.
Never ignore maintainability.
Flows aren’t always the answer — choose wisely
From your experience, what is one technical or architectural decision you see teams get wrong most often in Dynamics 365 projects or when creating solutions, and what would be a better approach?
It is definitely the most common worst decision to use Cloud Flows.
Microsoft has done a good job of making Power Automate understandable and very easy to use for creating new flows.
This is an architect-related problem. There should always be an architect who knows many areas of how requirements can be solved. Too often, I have had problems with flows where the logic runs pages and pages, which can be solved in 20 lines of a C# plugin.
Don’t get stuck in “what I know” or “what I like”.
Always go out of your cumfy box!
Your T‑shape: broad awareness, deep expertise
The Dynamics and Power Platform ecosystem evolves very quickly. What are your personal habits or systems for staying up to date with new features, best practices, and for remaining relevant long-term?
When I’m looking at the evolution of the Power Platform, I feel old…
15 years ago, I believe I knew pretty much all aspects of Dynamics CRM (by the way, D365 evolved into Power Platform, know that!). Now we have to know in your chosen area, probably a narrow area…
Someone wrote somewhere about the “T”. The horizontal means we should know about all areas, and the vertical means we delve deep into that specific topic. That is a good way to keep up with us; otherwise, we will drown in the firehose from Microsoft.
I do know a lot about Dataverse plugins and Fetch XML. I know little but many, many things, and I know who among my colleagues knows about the xyz area!
Go beyond using AI — understand it
For developers with 1–3 years of experience today, where would you recommend they invest their learning time over the next 12 months?
Unfortunately (sorry, I have to say “unfortunately”), I suggest you look at AI. Not just using it, but do that as well in Visual Studio and in VS Code.
More important is to delve into knowing how it is working.
I have a new car, running only on electricity, and I don’t really know how the battery takes me to the grocery store; I just accept that. It is annoying that I can’t explain it to my kid; it’s a bad position that I need to call someone if it doesn’t start. In some cases, we just have to live with it.
But with AI from a developer perspective, we can take control, or at least know how it is working for or against us.
Go test, go read, stay up to date with the coming versions, contemplate what might be good to use, and what we need to develop without AI.
Design your data model intentionally
If you could give one piece of advice to your younger self starting out with Dynamics / Power Platform, what would it be?
My main advice may be boring, and no one would follow it, but it is to learn well about databases, data models, and Dataverse. Designing the database well is crucial for a good solution and maintainability.
Go to a whiteboard, drawing models, start with which tables we need, which relationships, and which columns.
Options for tables, columns, and relationships are just as crucial.
The thing about normalization. Weird word? Bing it!
I know, all new or semi-new developers just want to go to vibe.powerapps.com or similar, see what’s a buzz this week, and let it generate a database and apps with a 2-second clickety-click, and tada, it’s done!
Think again.
Do my opinions help?
I’ve realised I carry around a bunch of opinions — but most of the time, they aren’t even formulated. It’s more like an itch. It’s usually only when someone asks that I stop, think, and try to figure out what I really mean. What starts as a vague gut feeling can (sometimes 😅) transform into something more understandable. And by doing that, I not only clarify my own thoughts, but also turn them into real, constructive suggestions that might actually help someone out there.
