Introduction
I don’t want to be the bearer of bad news—well, maybe just a little. My goal is to save you from yourself, so let me be clear: WE DO NOT CREATE UNMANAGED LAYERS IN PRODUCTION!
If you’re serious about maturing on the Power Platform, there are a few fundamental principles you need to grasp. I get it—the way Power Platform is set up, with everyone having a default environment, can feel like the wild west. But there are some non-negotiables. At the very least, you should be working with two environments—ideally, three: one for development, one for testing, and one for production-ready solutions. You know, a proper application lifecycle management (ALM) process—the same concept they drilled into us in college.
In theory—and in practice, based on what I’ve seen over the years (including my own mistakes)—you can do whatever you want. But at some point, you need to stop, take a step back, and ask yourself: Is a “hot fix” in production really worth the risk?
My answer? Absolutely not. And here’s why.
Governance
Yes, the G word—governance. Autonomy is great—until there’s a data breach. Then suddenly, we’re all asking how our favorite retailer leaked our PII (Personally Identifiable Information), and I’m staring at an email telling me to cancel my credit card, pack up, and start a new life.
Okay, maybe this article isn’t that dramatic—but it could be. Governance isn’t just a Power Platform concern; it applies everywhere. And while we champion secure backends like Dataverse, security only works if we take governance seriously. So, do yourself a favor—treat it like it matters. Because it does.
I have the privilege of working alongside some of the world’s premier experts in governance and compliance, partnering with some of the biggest names in business. And time and time again, we find organizations falling into one of two categories:
The Gentleman’s Approach
These organizations recognize the power of Power Platform but haven’t fully deployed it across their enterprise. They want to get ahead of the chaos by implementing governance before things spiral out of control.
Please Save Us
On the other hand, some organizations have already deployed Power Platform—without guardrails. Now, it’s out of hand, and they’re desperately looking for a way to regain control before things get worse.
If you’ve taken the Gentleman’s Approach, thank you. It makes life so much easier when we can focus on methodology and toolkits to set you up for long-term success.
If you’re in the Please Save Us camp, thank you as well. Acknowledging the problem and committing to steering your organization in the right direction is a crucial step.
The Problem with Poor Implementations
Here’s the thing—lack of governance in Power Platform leads to poor adoption. We have this powerful platform at our fingertips, but if security gaps, broken processes, or unchecked risks get in the way, no one wants to use it—or worse, no one can use it safely.
CIOs and the Tough Questions
Do CIOs make my job harder when it comes to governance? Yes. And they should. The tough questions need to be asked. Don’t just take my word for it—press me, challenge me, ask for sources. This is your organization, your investment, your future.
As a consultant, I care deeply about the services I provide, but at the end of the day, I will roll off your project at some point. My goal? To leave your organization better than I found it. And that is part of the reason we don’t create unmanaged layers in production!
The reasons that we’re here
Lack of Version Control
It’s right there, staring you in the face. You can do it. Just pop that flow open, add that missing action, and everything is right with the world—until it’s not.
Whatever the reason, you must operate with the mindset that tweaking production directly is not acceptable. You cannot predict unexpected behavior, and doing so completely undermines your version control and environment strategy.
If you have a production environment (and please don’t tell me it’s the default environment—that’s a conversation for another day) and some form of environment strategy, respect it.
Make your changes in the lower environments, test them, get sign-off, and deploy to production—the way it should be done.
Deployment Instability
In production, or at least in managed solutions within production, we have what are called solution layers. (I even included a nice, shiny image above to illustrate this—you’re welcome.)
When you create unmanaged layers, you introduce the possibility of deployment failures down the line. The artifacts and components in future deployments weren’t designed to handle the unexpected behaviors you introduced by manipulating a solution directly in production.
Save your deployments. Do it the right way.
Stick to the process—develop, test, get approval, deploy. Your future self (and your team) will thank you.
Difficult to troubleshoot issues
The degree of difficulty in troubleshooting can vary—but it becomes exponentially harder when you don’t even know what you’re looking for.
The unpredictability of removing a single component, a line of code, or an action in a flow can drive even the most rational person up a wall. One small tweak can have ripple effects you didn’t anticipate, turning a simple fix into a time-consuming nightmare.
Let’s avoid unnecessary cleanup, frustration, and sweat-inducing emergencies. Follow best practices, respect your environments, and don’t create an unmanaged layer in production.
Regardless of which camp you fall into, governance matters—whether you’re planning ahead or trying to clean up the mess.
Deploy managed solutions
This should go without saying—but I’ve seen it all. I’ve been on engagements where entire production environments were packed with unmanaged solutions, and it took months to convert everything to managed.
I want my clients to be set up for success, even if it means being a thorn in someone’s side. I get that introducing complexity, governance, and strategy might seem like extra overhead for IT. But trust me—it’s worth it.
Now, think about the alternative: years of unmanaged layers, uncontrolled deployments, and an unrestricted environment that you still call production. The overhead and nightmares that come with that far outweigh the effort of doing things the right way from the start.
If your goal is to build a mature Power Platform model, then you need a strategy in place. Create the necessary environments, security groups, and pipelines to establish a structured, secure, and scalable framework.
Call to Action
It All Starts with Education
Educate your business units, educate IT, heck—educate your dog if you have to. A mature Power Platform adoption is an organizational effort, not just an IT initiative. Do it the right way.
Use the Right Tools
You have plenty of tools at your disposal:
✅ The CoE Starter Kit – Provides a holistic view of your environment.
✅ Power Platform Admin Center – If you’re not ready for the CoE, at least establish a management strategy here.
✅ Security Groups (M365 or Entra ID) – Set them up early to enforce governance and access control.
One Rule Above All
Whatever you do, do NOT create unmanaged layers in production.