Another year, another release wave! Once again I dive into the documents and highlight my favorite features coming to the platform. Last time I did not do it and I have to say, I missed doing this. Diving into the release wave is something I do anyways. Making a blog post about it forces me to think a bit deeper about them as well. Another benefit is that I am sharing this with you, who hopefully enjoys reading my ramblings.
Features are rolled out to the Power Platform constantly now so the release waves mean less and less. However it is good practices to identify what’s coming, what’s relevant to you and what direction the platform is taking. For instance, this release wave mentions quite a bit about how Dataverse is preparing to scale what’s coming with agentification with everything build for Copilot Studio.
You can find the release plan on Microsoft Learn where you can also download the PDF or you can use the Release Planner. Often times features are already in preview and just finalized in this wave. In this article I will point towards the official documentation if available, else I will point to the Release Planner
Bigger features are also not announced through the release wave. They deserver their own bigger podium. However the Power App learn pages do mention 3 key investments. These investments definitely tickle my interest as one is quite new to me and the others are going to be incredibly powerful!
Key investments for Power Apps
- Plan Designer. Remember when we could ask AI to build us a canvas app? This is the same concept but on steroids. You can build entire solutions based on your input. Want to see it in action? Go check out this video from April Dunnam
- Extensible in app agents for Model-Driven apps. Already it was possible to add a copilot to your app, but now it’s possible to extend it using Copilot Studio! Bringing quite a bit of power and features to the Copilot. Again April Dunnam made an excellent video here.
- Bring your own agents. Not a lot is known about the implementation of this. But by the looks of it, instead of relying on the Copilot and Copilot Studio architecture, we can build our own.
Looking further through the Power Apps features there are a few that catches my eye. First I like to highlight some improvements to the offline capabilities of Power Apps. We can select columns for tables that are downloaded to the users device. These are added to the offline profile. This will help in performance as no longer all columns are in the dataset. Coming soon in preview and expected to be General Available in April.
AI is everywhere
The second is AI record summary to model-driven apps. Makers can define where this is enabled and what fields are actually relevant for the AI to summarize for the user of the application. This means you can tweak the AI to make it optimal for what your users want to see for such a summary field. To see this feature in action see this video from Reza Dorrani. I think this is a perfect example of how makers can utilize AI in a simplified manner to bring value to users.
Another example of this is Improve form fill assistance by adding custom knowledge. Now to be honest I need to see this feature in action first before I can truly understand this. We already have form fill assistance for users of our Model-Driven apps. But this feature will supposedly teach Copilot what fields need to be filled and with what. Something I will be keen to try out once this hits public preview in February.
Power Automates Copilot is going to be helpful!
Moving on to Power Automate Cloud Flows. I am incredibly excited about Create and edit expressions with Copilot. Creating complex Cloud Flows is not low-code. The expression language of Power Automate is so different then the rest of the platform it’s one of the hardest parts to get into. The amount of time explaining the language when I’m introducing new makers to the platform does not compare to Power Fx. Having a copilot to help you build and explain formulas is exactly what we need!
The next feature I want to highlight that I still want to play with is Create generative actions in cloud flows. The ability to create intent based Flows instead of hardcoded flows intrigues me. As you might know I have a developer background and it’s quite hard to grasp some of the AI features, but this actually makes sense. Abstracting hardcoded mapping and offload it to AI and describe a desired outcome sounds like a good solution to the unstructured input problem. I am curious how pricing will change as we see more and more of these AI features integrated into the product.
Talk of the Town: Copilot Studio
On to the Copilot Studio features in this Release Wave. I am most excited about the ability to Discover and install Agents from Copilot Studio. This functionality will bring Copilot Studio Agents to the Power Platform Catalog where other makers can install them. After installation makers can extend the agents, or just look at how others build theirs. Admins will have granular control over which Copilot Studio Agents will be shown to which environment. By the look of things, Microsoft will even publish a few of their own!
Another feature that I like is the Add suggested prompts to custom copilots. Microsoft 365 Copilot Chat is becoming one of the primary interface where users interact with AI. It’s beneficial that Copilot Studio agents act the same way. Get suggested prompts to highlight what your Copilot does best. I’m curious to see how these suggested prompts will be placed in for example Teams or Webchat.
Last but certainly not least I want to group together 3 features:
- Analyze knowledge source usage for custom agents.
- Test and debug agent actions in Copilot Studio.
- Analyze action usage for agents.
These features will be incredible helpful in monitoring what brings the most value to your users and optimize them. Both on an iterative basis when developing, or when you already have a Copilot Studio Agent deployed for your users.
These are my highlights for Power Apps, Power Automate Cloud Flows and Copilot Studio. I hope my small little insights are of use to you, and if you made it this far, thanks for reading!