Cameron Steele, CEO, Prophia
Turning Lease Documents Into Intelligence with AI
Subscribe now on any of these platforms so you don't miss a single episode.
Apple Podcasts
Blubrry
Android
YouTube
TuneIn
Stitcher
Google
Spotify
Amazon Music
RSS
Guest: Cameron Steele, CEO, Prophia
In brief:
-
Lease documents remain the primary source of economic truth in commercial real estate, yet most of their critical data remains trapped in unstructured legal documents.
-
Many CRE firms still rely on manual processes or outsourced labor to extract lease information.
-
AI now enables lease data to be converted into structured intelligence that can support operational decisions and portfolio management.
-
The greatest barrier to adoption is not technology but organizational change management.
-
Firms that fail to modernize their data infrastructure risk operating with incomplete or inaccurate information.
The hidden infrastructure of commercial real estate data
Commercial real estate is widely described as the world’s largest asset class. Yet many of the industry’s core operational processes remain surprisingly manual.
In my conversation with Cameron Steele, co-founder and CEO of Prophia, the problem becomes clear quickly. Despite billions of dollars in property assets under management, the economic blueprint of those assets often resides in scattered legal documents rather than accessible data systems.
Steele describes leases as the essential informational foundation of any commercial property.
“Leases are a blueprint of value for buildings. So if you want to understand how to value a commercial building, you need to look at the leasing data.”
That blueprint contains the economic terms that determine revenue, risk, and operational obligations. Yet historically, extracting and managing that information has been labor-intensive.
The persistent manual workflow behind CRE operations
The core issue lies in how lease information is handled.
Key terms such as rent schedules, renewal rights, expansion options, and operating obligations typically reside in dense legal agreements. Extracting that information has traditionally required analysts or outsourced service providers to read documents and summarize them manually.
Steele notes that this process is still surprisingly common across the industry.
“Everything we do is driven off of the documents and the structured data we pull out of the documents. So it’s all done manually today.”
This manual workflow introduces several risks. Data can be inconsistent between systems. Important provisions can be overlooked. And critical decisions may rely on summaries rather than the underlying agreements.
These challenges become particularly acute for mid-sized real estate companies managing complex portfolios with limited operational staff.
Why structured and unstructured data must work together
The technological opportunity lies in connecting two types of information.
First, there is structured data - the extracted fields such as square footage, lease start dates, rent escalations, or termination rights.
Second, there is unstructured data - the underlying legal language in lease documents themselves.
Historically, commercial real estate software has focused on structured databases. The nuance of the actual documents remained outside the system.
Steele argues that modern AI tools allow both forms of data to be integrated.
“We think about data both unstructured - so documents - as well as structured, which is data we pull out of documents. We tether and integrate the two elements so they’re there together.”
This integration matters because lease terms often contain nuance that cannot be captured fully by a simple database field. The underlying documents remain the definitive source of truth.
Where AI changes the equation
Artificial intelligence enters the workflow through document processing and query tools.
Large language models and natural language systems can identify relevant terms inside lease documents and convert them into structured datasets. Human verification then ensures accuracy at an enterprise level.
Steele emphasizes that automation alone is insufficient without reliability.
Customers require what he describes as “enterprise grade 99% accurate verification” when working with financial contracts.
The result is a system where users can query both extracted data and the original documents.
Instead of manually reading hundreds of pages of leases, users can ask targeted questions about tenant obligations, termination rights, or portfolio exposures and receive documented answers linked directly to source material.
The operational questions real estate firms constantly ask
For asset managers and portfolio executives, the value of this approach becomes clear when examining common operational questions.
Examples include:
-
What maintenance obligations fall on tenants versus landlords?
-
Which tenants have upcoming termination rights?
-
What encumbrances affect a specific space?
-
Which leases contain renewal options?
In traditional workflows, answering such questions often required days of analysis.
According to Steele, the goal is to compress that timeline dramatically.
“We’re trying to just give you tools that give you information very quickly.”
By linking queries directly to both structured fields and underlying documents, AI-enabled systems can produce answers within seconds while preserving auditability.
The real barrier: change management
Technology, however, is rarely the limiting factor.
The primary obstacle facing many CRE organizations is organizational inertia.
Many companies have relied on the same operational processes for decades.
Lease abstraction, document management, and rent roll compilation often follow procedures established long before modern data infrastructure existed.
Steele frames the challenge bluntly.
“I think it’s change management, honestly. People have been doing something forever.”
Legacy vendor relationships, established workflows, and internal habits can slow adoption even when new tools offer clear efficiency gains.
The economic implications
Beyond efficiency, better lease intelligence can also uncover unexpected financial value.
During onboarding for one large office asset, Prophia discovered a licensing agreement referenced in lease documentation but missing from the owner’s rent roll. After reviewing the agreement, the owner invoiced the tenant for nearly a decade of unpaid licensing revenue.
Such examples illustrate how incomplete data systems can obscure material economic details.
For many owners, improving data visibility becomes not just an operational upgrade but a financial one.
Why data infrastructure may become a competitive advantage
The broader implication for commercial real estate is that data infrastructure is evolving from administrative support to strategic capability.
Firms able to quickly understand lease exposure, tenant rights, and portfolio trends may operate with a clearer understanding of risk and opportunity.
Conversely, organizations that continue relying on fragmented data processes may struggle to keep pace as AI-enabled tools accelerate decision cycles.
As Steele puts it, the industry historically invested heavily in buildings and people but not in data.
That imbalance may now be shifting.
Bottom line
Commercial real estate has long relied on legal documents as the ultimate source of truth for property value and risk. Yet the operational systems used to interpret those documents have lagged behind other industries.
AI-driven data platforms are beginning to bridge that gap by converting lease agreements into structured intelligence while preserving the underlying legal context.
For CRE professionals responsible for managing complex portfolios, the implications are clear: the ability to interrogate lease data quickly and reliably may become an increasingly important operational advantage.
This discussion is particularly relevant for asset managers, portfolio executives, and multi-asset sponsors navigating growing complexity across their portfolios.
Related to this episode:
