"We started our AI journey by focusing on governance first," Joseph Martino stated, as several executives around the room raised their eyebrows. It wasn't the flashy AI starting point many had expected. "That means formal policies, committees, and a modern data framework. AI models are only as reliable as the data you feed them."
As Vice President of Information Technology at Primaris REIT, Martino brought a refreshingly pragmatic perspective to our CIO ThinkTank Series on AI & Business Transformation. While many organizations rush toward AI implementation, Martino's approach addresses a sobering reality: 30% of enterprises cite lack of governance as the top barrier to scaling AI, according to IDC's 2025 research.
In an era where everyone wants to talk about generative AI and large language models, Martino focused instead started panel discussions on the foundation that makes these tools truly valuable.
For those interested in diving into the other articles in this CIO ThinkTank series they have been listed here for your convenience:
Read on if you would like to learn more about getting the balance right between experimentation and strategic governance.
From Data Silos to Data Fabric: Creating an AI-Ready Architecture
The room was quiet as Martino shared how Primaris REIT had transformed its data architecture from isolated departmental databases to a more accessible "fabric" approach—a strategy aligned with many industry leading frameworks.
"In the real world our data is often siloed. As an example for some organizations their property management data, tenant information, and operation systems property descriptions in emails live in separate worlds," he explained, drawing a mental image for the executive audience. "Our first months weren't spent on AI pilots—they were spent creating a unified data ecosystem where AI could eventually thrive."
This data transformation paid immediate dividends, even before adding machine learning capabilities:
"We reduced reporting time significantly," Martino noted. "Analysts who previously spent a large portion of their time gathering data could now focus on deriving insights from it."
Microsoft’s own implementation of a unified data platform using data fabric principles resulted in a 50% efficiency boost, showcasing how such strategies can streamline workflows and reduce time spent on repetitive tasks like data preparation. Research from many others supports this approach, showing that companies adopting a unified data fabric strategy see dramatically accelerating AI project timelines.
"When we cleaned up our data flow and integrated privacy labels, it was like removing bottlenecks," Martino emphasized. "The AI committee could actually focus on building solutions, not untangling messy data."
The Two-Committee Approach to AI Governance
Recognizing the complexity of AI implementation, Martino encouraged others to leverage industry frameworks and adapt it for AI Adoption as almost all of them have already done the work for you. Yet it was his explanation of how his organization approaches a team based model around governance that resonated strongly with other executives at the panel and throughout the room.
Together they established a multiple team and dual committee structure that mirrors several organizations in attendance:
"We keep these two committees aligned, but distinct," Martino explained. "If a single group attempts both strategic alignment and deep technical vetting, neither function gets the attention it needs."
This approach can pay dividends as regulatory requirements evolve. The upcoming EU AI Act mandates robust governance for "high-risk" AI systems, and forward-leaning companies like are voluntarily adopting these standards ahead of regulation.
"Being proactive about AI compliance is a competitive advantage, not a burden," Martino asserted. "When regulations tighten—and they will—we'll already be operating at that higher standard."
Martino cited the value of Microsoft Purview, Cloud App Security and other capabilities that help organizations assess, understand and protect their data in a scalable way. “The AI Hub experience that shows you how people are using tools today when you use Defender? It’s invaluable and makes getting started much easier.”
Microsoft Purview should be a part of any Microsoft customer’s comprehensive data governance program but it can also provide risk mitigation for generative AI in the short term while the broader governance program is underway. The risk mitigation Microsoft Purview provides for AI can begin on day one. This includes Microsoft AI, like Microsoft 365 Copilot, AI that an organization builds in-house, and AI from third parties like Google Gemini or ChatGPT.
Getting the strategy and rollout for AI Adoption isn’t always easy and tackling oversharing and AI volume challenges requires the right approach. There are proven approaches like the 10-step strategy we covered in our webinar to close the gap in data and content governance. For more insights like this consider exploring another popular article series on tackling the waves of AI transformation where we break down how to achieve sensitivity and lifecycle management that scales in the era of AI.
The ROI Question: Making AI Pay Its Way
When the discussion turned to budgeting for AI initiatives, Martino shared a pragmatic approach to managing costs and proving value. By running short, targeted proofs of concept before full implementation, his team validates actual ROI and avoids scenarios where the organization invests in expensive solutions that never scale.
This approach to cost-effectiveness is also reflected in the work Paragon Micro does, helping companies like Primaris REIT effectively manage the financial aspects of AI adoption. By assisting organizations in evaluating AI tools, and optimizing data infrastructures, Paragon Micro ensures businesses can scale AI without overspending or compromising on performance. Their expertise in AI integration and governance helps organizations maximize the value of AI while staying within budget.
“Let’s say a pilot reduces lease processing time by 42%, that creates a clear business case for expanding the implementation to more properties in our portfolio.” Focused surveys by McKinsey show that organizations effectively transitioning from pilots to scaling prioritize data-driven evaluations of pilot outcomes, including ROI metrics, which directly influence their ability to expand initiatives.
"We tie every AI project to specific KPIs," Martino explained. "For property management, that might be maintenance response time or tenant satisfaction scores. For finance, it could be invoice processing time or forecast accuracy."
Microsoft Copilot: A Test Case for Enterprise AI Value
Martino shared how Primaris REIT is carefully evaluating Microsoft 365 Copilot for potential productivity gains across the organization.
"Rather than immediately deploying Copilot enterprise-wide, we identified departments with different use cases," he explained. "Our leasing team could use it to draft and refine property descriptions. Finance could use it to analyze lengthy contracts and extract key terms. And our executive team could use it to summarize meeting transcripts and action items."
This targeted approach allows them to measure concrete benefits and get targeted feedback before expanding—a strategy that's proving effective as they build the business case for broader rollout.
Future-Ready Data and AI Governance
As real estate continues to evolve, Martino is always working as a leader to understand the right opportunities his organization should explore, whether they be leveraging AI and AI agents to optimize tenant experiences or orchestrating "smart building" systems in real time.
"Our proactive approach to data architecture means we can now layer in AI capabilities like predictive maintenance or space utilization analytics," he noted. "Because we built the right foundation, we can implement these innovations much faster than competitors starting from scratch."
Martino concluded with a perspective that resonated across the room: "As new regulations and technologies emerge, it's the discipline of governance and good leadership that keeps our AI efforts aligned with our core business values. ThinkTank events like this are a prime example of a place to share those frameworks—so we all accelerate responsibly."
As the panel ended and throughout the day executives crowded around Martino, eager to learn more about his successful committee structure and governance templates—evidence that sometimes the most valuable insights come from the least flashy aspects of AI implementation.
In our next ThinkTank article, we'll explore how one CISO is leveraging AI to transform cybersecurity operations in response to increasingly sophisticated threats: Defending at Speed: The Security Imperative for AI
As AI continues to evolve, responsible governance remains key to its success. To explore this crucial topic further, we invite you to join our upcoming CIO ThinkTank webinar: AI Agents at Work: Driving Value While Governing Risk. Learn from experts as they discuss strategies for driving value and ensuring AI initiatives align with your organization’s core values.
Join Our Mailing List
