Weekly Briefing

article sponsor image
Partner Content

Rolling Out AI in Commercial Real Estate: The Five Pillars

May 13, 2026 | Brian Sutherland, VP, Sales, Yardi
5 min read
listen to article Listen to this article

Something has shifted in the conversations I'm having with CRE technology leaders. A year ago, the question was "what is AI going to do to our industry?" Today it's "how do we actually roll this out?" That's a meaningful change. The curiosity phase is over. What comes next is harder – and more important.

At Yardi, we work with some of the largest real estate owners and operators in the world. Over the last 18 months, we've moved from theorizing about AI to deploying it inside the workflows our clients run every day. What we've learned is that the firms getting real results aren't the ones chasing the most sophisticated technology. They're the ones taking a prescriptive, deliberate approach: starting with clear wins, building trust in the technology and expanding from there.


Here's the framework we've developed for how CRE organizations should think about their AI journey.

  1. Start with the work you already own.

    The most common mistake I see is organizations going outside their existing platforms to find AI solutions before they've fully leveraged what they already have. Your core property management and accounting system sits on top of your general ledger, your payables, your receivables, your leases, your vendor relationships. That data is extraordinarily valuable and agentic AI can work directly within it.


    Before evaluating third-party AI tools, ask your largest software providers what's already available natively. The wins are closer than most teams realize. Agents that automate invoice approvals, scan leases for billing discrepancies, or handle month-end workflows don't require new infrastructure. They run inside the system your team uses every day, with the permissions and governance already in place.


    First wins build organizational confidence. That confidence is what allows you to scale.

  2. Think like a service provider: segregation of duties applies to AI, too

    Commercial real estate organizations have always operated with clear segregation of duties. The person who codes an invoice doesn't approve it. The analyst who builds the model doesn't make the investment decision. That discipline exists for good reason with regard to fiduciary accountability, audit readiness and risk management. AI doesn't change that principle, it extends it.


    Think of agentic AI the way you think about your outsourced service providers. A full-service payables team handles specific, well-defined tasks within a governed process. They don't have unconstrained access to your systems. They have scoped access with clear responsibilities and audit trails. Agentic flows should work the same way. An agent handling AP approvals operates within defined thresholds, flags exceptions for human review and logs every decision. An agent scanning leases for unbilled charges surfaces findings for your team to validate and act on.


    This framing matters because it makes AI legible to your board, your auditors and your risk team. It's not a black box; it's a governed participant in a process you already control.

  3. Focus your agents on three outcomes: risk, cost and time

    Not all AI use cases are created equal. The ones that earn organizational trust fastest are the ones where the outcome is quantifiable and the stakes are clear. In our experience, the highest-value agentic flows fall into three categories.


    Risk reduction. Lease audit agents that scan every lease in your portfolio for billing that isn't set up correctly. The revenue leakage from incorrectly configured charges across a large portfolio is often in the millions and it's largely invisible without AI doing the work at scale.


    Cost savings. Vendor terms agents that scan invoices across your entire vendor base for early payment discounts that aren't being captured. Most portfolios have early payment terms sitting unconfigured in less than five percent of vendor records. An agent can find them, configure them, and quantify the NOI impact before you've made a single decision.


    Time savings. Approval automation that handles the majority of routine invoice approvals without human touch frees your team to focus on exceptions that actually need judgment. For large operators, this represents multiple FTEs of recovered capacity per year.


    When you evaluate an AI use case, start by asking which of these three buckets it falls into. If the answer isn't clear, the use case probably isn't ready for production.

  4. Data access is a competitive advantage: govern it like one

    One of the most significant developments in enterprise AI over the last year is the emergence of Model Context Protocol, or MCP. If you haven't encountered this term yet, you will. MCP is an open standard that allows large language models: Claude, ChatGPT, Copilot and Gemini, to securely connect to and query live operational data in real time. Think of it as a governed pipeline between your AI interface and your systems of record.


    For CRE, the implications are significant. Your property performance data, your financials, your leasing pipeline and your work orders can all be surfaced conversationally, without manual exports, without stale reports and without someone pulling data into a spreadsheet first. An asset manager can ask a natural language question about portfolio NOI variance and get an answer drawn from live system data in seconds. A leasing team can pull a rent roll, layer in market comparables and model scenarios, all inside the AI tool they already use.


    The key word is governed. MCP doesn't mean open access. The protocol allows your software provider to define exactly what data is exposed, to whom, and under what conditions. User-level authentication, property-level security, read-only constraints and full audit trails on every query can and should be built into the connection. The organizations that will benefit most from MCP are the ones treating it as a strategic data access layer with clear controls, not a shortcut around their security posture.


    Security and governance aren't the cost of doing AI right, they make real-time data access safe enough to actually use.

  5. Partner with providers who have done this at scale

    The final pillar is arguably the most practical. CRE technology leaders are being asked to make consequential decisions about AI at a moment when the landscape is moving faster than any individual organization can track. The firms I see navigating this most successfully aren't building from scratch, they are leaning on their largest software partners to bring a prescriptive point of view.


    That means expecting more from your vendors. Ask them what they've deployed with clients that look like you. Ask them what the ROI calculation looks like before you sign

    anything. Ask them how they handle the governance and security questions that your IT and legal teams will raise. The right answer isn't a roadmap slide, it's production deployments, reference clients and quantified outcomes from real workflows.

Applied AI in commercial real estate is no longer speculative. The organizations treating this as a change management challenge rather than a technology evaluation, and partnering with providers who've already navigated that terrain, are the ones that will be positioned to lead when the market rewards speed and intelligence. The window to build that foundation deliberately is still open. But it won't be for long.

Brian Sutherland, VP, Sales, Yardi
Brian Sutherland leads Yardi's commercial sales and marketing organization, overseeing go-to-market strategy across office, retail, industrial, coworking, self-storage, manufactured housing, and corporate real estate. As executive sponsor for Yardi Virtuoso, he helps property and asset managers move from AI curiosity to AI execution across their daily workflows. Before Yardi, Brian spent 15 years founding and leading startup companies.

This Week’s Sponsor

Yardiฎ develops and supports industry-leading investment and property management software for all types and sizes of real estate companies. Established in 1984, Yardi is based in Santa Barbara, Calif., and serves clients worldwide. For more information on how Yardi is Energized for Tomorrow, visit yardi.com.