his three-phase approach to enterprise copilot strategy begins with establishing a strong foundation in security, data integrity, and AI adoption. It then shifts to building targeted, high-value AI and agent solutions at the business-unit level. Finally, phase three brings these solutions together through an orchestrator agent, creating a unified and cohesive enterprise copilot experience without agent sprawl.
Hey everyone, this is Venice, and in this video I wanted to share our phased approach for enterprise co-pilot growth strategy and answer a question that I'm hearing a lot from our customers that I also had myself. So let's go to the visual.
Okay, phase one. This is where most of our customers are at right now. In phase one, we're laying the foundation for AI and co-pilot growth strategy. What this looks like is looking at our security policies, looking at our data integrity and data strategy, and then laying the foundation upon which Microsoft 365 Copilot can succeed.
So building an AI culture and really honing in on adoption and training. A lot of customers also start with an enterprise-wide use case for building their own co-pilot in this phase. And a lot of what we see are the frequently asked questions from an HR policy use case—something that is not super complex to build and provides value to the whole enterprise so that we can use this as a win and start getting adoption of the use case around why we would want to build our own co-pilots in the enterprise.
So these things lay the foundation for us to move to phase two.
This is where we look to see strategically where it makes sense to invest more time and money into building our own AI and agent solutions in the business. Most often these are not enterprise-wide use cases but more specific to the business level, where there are things that we should and can automate that would provide a lot of value for a smaller group of people.
We've talked about this in our other videos, and I'm happy to dive more into the strategy and approach of how to go about identifying those things and what that can look like. But know for phase two it is identifying certain solutions for different areas of our business, building, and iterating on that.
This is where the question comes in: how is this going to grow? Is this the end goal? Is this going to be a sprawl problem where now I'm going to be asking my business to go to this chatbot for this and then to switch over to this chatbot for this? It feels really disjointed, and is this really where it's at—that I'm just building a bunch of different bots?
This is an important question to know that really we're looking to get to phase three, where we're going to create an orchestrator agent that we can use to create a unified communications layer across our agents to bring them all together. It doesn't make sense to do that initially, because in phase two we're building the value of these agents, where phase three is only valuable with the phase two value there.
And we're wanting to connect them and build on the security from phase one and the value from phase two to bring it all together so that we're delivering an enterprise co-pilot experience that is cohesive and tailored—where people have access to the AI and agents that are for them and their business level, but they are not switching between different co-pilot agents. We're delegating that to the orchestrator agent.
So that is the big picture, and I think a lot of people are missing phase three, where this can go, because we're working on phase one laying the foundation, we're working on phase two mapping out where we're going to go next, where the ROI is for a business-specific solution build, but ultimately we're going to bring it all together in phase three.
I hope that helps. I would love to discuss what this looks like in your business and any other thoughts that you have.
