A (5:15)

So thank you, Jim, I appreciate that.

B (5:19)

Absolutely.

A (5:19)

I really like how you started with just kind of outlining, you know, starting out with the business outcomes. You know, you need things that are measurable and you're looking at things that can provide value, true value. And I know for your, your customers, you design for value and you build in that resiliency, as you mentioned. But I'll just highlight a couple other things. Good governance, track the ROI and have a good roadmap. So I really appreciate that. And Jim, the next question I have for you, you developed an AI adoption playbook at ProArc, especially around Microsoft 365 Copilot. What are the key elements or steps in that playbook that enable organizations to turn AI hype into practical working workflows? Maybe you could walk us through an example use case where this playbook delivered measurable business value.

B (6:06)

Sure, absolutely. You know, like anything else, the use of generative AI, or in our case, you know, more specifically copilot, we treat it like any other enterprise initiative. So you have to make sure that you start with a secure foundation you look at and try and target some very high value use cases that can establish success early on. It's also very important to make sure that you identify the folks within your organization who are very excited about this technology. You call those our champions. And then you have to measure relentlessly to verify that you're actually getting what you expect. We call this approach our Smart Start approach. And it really starts with looking at your data estate. So we're talking about Copilot specifically. You know, we run a Microsoft 365 data risk review, go through the entire data graph data and also do a gap analysis. We're looking at things like identity, we're looking at how the M365 environment is configured. We're looking at how data labeling and sensitivity, labeling and protection is done. Something that most organizations aren't doing enough of. And the purpose of that is we want to make sure copilot only sees what it should or what you want it to see. As I mentioned before, we define those first wins. We may pick two or three Personas or roles in the organization. This may be sales or project management or finance. We build out a roadmap and we assist them by we're kind of aligning to Microsoft's adoption guidance, if you're familiar with that. But the guidance is get ready, you onboard, engage and then you deliver the impact. One thing you can overlook though is Standing up the human system. We think of AI as kind of taking away all these responsibilities, but there still are humans in the loop here. And so we need a good AI usage policy that can establish the ground rules for those users. But we also very strongly believe in standing up knowledge hubs. And I call those knowledge hubs centers of excellence where your users can interact with your champions to get very specific role based training to their position and their work, as well as prompt toolkits that they can use to really give them some early confidence. Because there's still a lot of skeptics out there and they'll lean on AI and then they'll kind of tailor a tail off and then they'll kind of come back to it when they're, when they're asked to. But it's important that they constantly are being, this stuff is being reinforced and very important that we measure and iterate. So as the person who leads adoption and not my organization, we use things like my M365 admin reports and copilot dashboards to look at our adoption score, make sure that we can actually demonstrate the cost and value to the organization. As far as assisted value, we can actually translate into dollars and cents, which is very powerful as it relates to a specific use case that we found a lot of traction in. And we've built a lot of copilot agents. But one specific agent that has really resonated very well is we built one called a Microsoft Funding Finder agent. And that agent is used by our sales teams to help them unlock or find potential funding that we can apply to our customers to help them in their modernization journeys. If anyone has dealt with Microsoft over the years, and we're a very deep partner of theirs, you'll know that it's hard to wade through all the information. So we kind of brought all these resources together in one location where our account teams can ask a question, give the context of what type of project we're working on with those customers and see if there's something that matches that customer to be able to apply those funds to that, that engagement. And that has really dramatically reduced the amount of paperwork. It's helped us win more deals on and on beyond the agents. We're able to again, as I said before, demonstrate hundreds of hours a month of copilot assisted work across our users, which typically can translate into tens. Well, right now I think we're tracking around $15,000 of assisted value every month across the hundred users or so that we have licensed. So, you know, that's that's kind of where, where we really, that's our approach to, to AI overall.

A (10:29)

Thank you. I appreciate you unpacking all that. I know starting out with a project like this, it's. You start with that secure foundation, you establish success early on where you can see those early wins. You're constantly measuring, iterating, of course, and then you build out that roadmap that you can actually follow along the way. And then of course, building out knowledge hubs, as you mentioned, to ensure rollout success. I liked your example of that agent. I think it's Microsoft Money Finder, I believe you said.

B (10:54)

So, yeah, ultimately, yes, it's called Funding Finder, but we should have called it Money Finder.

A (11:01)

No, that's great. I appreciate that. So that's, that's just awesome. So, Jim, next question. You stress that technology isn't just about systems, it's about, about people. How do you help organize? How do you help organizations, CIOs, IT leaders build, buy in, change culture and prepare teams for AI and cloud transformation, especially for those resistant to change or maybe wary of automation.

B (11:24)

Yeah, it's not easy. You know, there's a lot of, a lot of organizations understand that they're being expected to do something. They're, they're seeing a lot of their peers and other organizations, but they're just unsure. So really you want to start with safety first and it's proarch and we encourage this in our clients as well as we, we actually have an AI governance council and we also make sure that we have an AI usage policy that's defined very early on. So this kind of reduces the anxiety for those users about, you know, what's permitted, what are the norms and also signals to them. And it's actually part of our, our mission statement as the governance council, that AI is not here to replace them. Right. It's here to make them better at what they do. It's here to augment their capabilities. So this ultimately anchors the enablement and the risk tolerance and compliance. If you think about the industry of robotics, for a long time they were all about the 3Ds. I don't know if you're familiar with the 3Ds, but it was get rid of the dull, dangerous and dirty work. They're similar in the AI world where getting rid of the dull, disruptive and draining work for our users. And by doing that, you can actually show them immediate, immediate relief. And they kind of understand, wait a minute, this tool is here to help me. It's here to take meeting notes, which I find onerous. It helps me get through those long email threads that I have to summarize after I've been on vacation for a long time. It gets you started in those initial drafts of documents and it's amazing how fast the skeptics can fade away when someone realizes they can get a power back in their day to do things that are more creative and more productive. And we, we run this play openly within proarc. We also, like I said before, we build a champions network and we train in the flow of work and that's very important. We don't make this kind of an option. We say this is the new way you're going to do work and this, these are the tools you're going to use and this is the process you're going to use. And so it's important you identify the early adopters. We have a user group that we run every month where we're constantly training folks on what's the latest and the greatest things within AI and Copilot and we share what's working and what's not working. As I said before, we really adhere to that adoption framework from Microsoft. We can make this repeatable as we continue to expand it across the organization or when we're working with our customers as well. And you know, it's, it's, it's really about, you know, what's going to stick with those users and being able to prove, you know, return on investment to the, to the exec teams.

A (14:00)

Thank you, that's awesome. You know, I really like how you take this approach. You know, obviously there's a lot of anxiety, ambiguity in this. Right. How do you reduce that? Well, you have to have some structure in place. Like you very familiar with the AI Governance Council's AI Usage policy. And again, taking my, that messaging of taking away the fear that these tools are here to help and augment what they're trying to do. I like how you highlighted reduce that dull, disruptive and draining work. I think that's important. Get that message out there but really identify those early adopters, capitalize on that success. Appreciate that message. And Jim, the last question of the day. What developments in cloud, AI or adjacent areas, maybe edge computing, large language models, MLOps do you believe will be the most disruptive over the next three to five years? How should organizations begin preparing now so they aren't caught off guard?

B (14:51)

And I think there's a lot of things that are obvious that are going to happen and then there's a lot of things that we just don't know yet. Right. And so, you know, on the obvious front, agent, agentic AI is definitely here to stay and it's going to be built into the enterprise stack. You look at what Walmart's doing right now, you know, they, they started out with all these various agents and now they kind of this one monolithic agent that runs all of their business. So we're going to be really seeing a shift from copilot as a personal assistant to actually processing work that spans teams and systems, not just for individuals. So think of it as, you know, forecasting what's going to happen in the business, helping on the customer support front, being able to provide better security coverage. So the other thing too, I think you're going to really, what we already are seeing is around SecOps security operations, Microsoft has their platform for Copilot for security. You're going to see a lot more of that built into systems. First party, third party where AI is going to be used to triage threats or vulnerabilities or incidents and respond in minutes and not hours. Instead of have to have humans looking at this stuff constantly, it's going to alert them to the fact that okay, we found a threat, we took care of the threat, this is what we did. But even with those more sophisticated things, being a partner in that, troubleshooting or hunting to kind of try and pin down where the vulnerability or threat is coming from. And then if we look at data governance, this is something that a lot of organizations have struggled with tremendously, but it's not really going to be an option as we look at the age of AI where data is not that it's not important now, but it's really important to have good data governance and, and also treat your data as though everything has a purpose, everything should carry a label, everything should be protected based on what that content is and making sure all of that's in place. And many organizations are very much behind in all of that. And so, you know, we're definitely here to help our customers, but it's something that's new and it's going to take some time. And then you know, if you look at what are the things to, for organizations to be aware of so they don't get caught off guard when all these changes are happening. As we mentioned at the beginning of this, this discussion, make sure you're standing up those landing zones, establishing good governance for your cloud environments, ensuring that there's the right rollbacks, role based access controls in place from the very beginning, not discounting Resiliency. It's great to use the cloud for a lot of different things because again, you can scale very well. It's very dynamic. You can scale in, scale out, you can build things very quick. But you have to make sure that, that they're resilient. And you know, around AI, one of the things that I think that a lot of organizations overlook is this concept of a governance council because they really are just getting started and they don't even have kind of a sense of well, what that, what would that look like? But once you get people in the room and you start having conversations, you realize that now more than ever you can't really get away from not having eyes on how this technology is being used and all the right players in the room. So for instance, our governance council, close our ciso, it includes our representative from legal, people from our digital engineering teams, people from our HR teams because they all need to have input into how these technologies are going to be used and how they're going to be used safely and ethically. And lastly, really it's about making sure if you're tinkering with AI, you know, start thinking about where the business pain points are. Don't come up with use cases first, but think of what the pain points are, what you're trying to solve for and pilot two agent two use cases and tie them to explicit KPIs that you know are going to be able to demonstrate value quickly and then scale that as you find things that work.

A (18:49)

Thank you, really appreciate that. There's so much there and, and you know, as you look ahead, you'd mentioned there's some things that are obvious, some things that are not. But you know, agentic AI is being built into the enterprise stack today. But as you mentioned probably see a gradual move from the copilot personal assistant to those enterprise agents which I see organizations doing that today. You mentioned SecOps. I think real time monitoring, alerting and actionable tasks to remove threats are important. Again, with so much data and so much, so many endpoints to protect, I think it's important to leverage AI there and establishing good governance and role based controls and access. I think it's very important as well to help keep things kind of buttoned down and then building that resiliency. I really think what you talked with our audience today is very helpful and it's going to help a lot of folks start to navigate down their roadmap. So I really appreciate that. And Jim, it was such a pleasure having you on today and I look forward to speaking with you real soon.

B (19:47)

All right. Thanks, Brian. Appreciate it.

A (19:50)

Bye for now.