sponsored by:
MRI Software

What can the AI-enabled future of real estate investment look like?

By Vijay Anand and Arik Kogan

As every industry races to incorporate more artificial intelligence features, experts are hyping the ways this technology will transform business across every organizational level. Still, something is missing when it comes to real estate technology – especially for investment firms.

Most conversations focus on how time-saving automation reduces operational costs and boosts profitability for property owners and operators. The industry hasn’t yet realized AI’s full potential to improve the daily lives of asset managers. Developments are coming – and already happening – that go beyond simple task automation.

We’d like to share some predictions of how AI will help asset managers focus on the higher-level, value-creating parts of their jobs in the next few years. You’ll learn how scenario modeling will incorporate formerly unavailable data without complex programming skills, how portals can turn investor questions into insight-filled reports, and how asset and portfolio managers can focus on delivering value while boosting investor confidence.

What is the AI-enabled future of asset management? Read on to learn more.

    Don’t wait for the future. Get the investment tech you need to win today.






    * By clicking the button below you agree to have your information shared with our sponsor.

    Leveraging new data and increasing analytical capacity

    Expanding the data universe

    Real estate is no stranger to data analytics. Properties generate incredible volumes of data daily, from occupancy rates to energy use to foot traffic. When considering new opportunities, asset managers must examine this data as well as the current regional, industrial, and economic picture. With multiple deals in play, the amount of data on one person’s plate gets out of hand quickly.

    Take energy use, for example. Accurate evaluation requires researching not only a property’s utility costs over a defined period, but also the utilities used, contributing environmental factors, and the costs for similar properties – and that’s before cleaning the data. Now multiply that workload by all other comparison properties, and then multiply that by all potential investments an asset manager could make across different portfolios. Deciding where to invest capital becomes overwhelming.

    Rather than let deals languish or slip away, asset managers can use artificial intelligence to find, load, and analyze vastly more amounts of data. Freed from manually hunting asset- and macro-level information, they can evaluate more deals across more regions and asset classes and update underwriting models without impeding investment timelines.

    For example, AI-driven systems like MRI Agora Insights could aggregate market-level data from commercial, residential, and industrial properties with more details like lease obligations and facilities management costs. From there, intelligent systems could quickly analyze the datasets and uncover multiple correlated metrics. Best of all, asset managers can use natural language processing capabilities to get the insights they need through simple queries. An asset manager could ask questions like “What is the most recurring charge within the New York area?” or “List the top ten occupants who have increased deals this year by over 20% across the portfolio.” AI can examine thousands of lines of transaction data and more to not only find the answer, but also generate reports – helping a lone asset manager complete 3–4 hours of research and analysis in minutes by asking conversational questions.

    Taking advantage of new data types

    AI will also help asset managers incorporate more data types and sources into valuation models. Advanced analytical models can uncover correlations between diverse, disparate data and enable broader perspectives too difficult for an asset manager to do alone. In addition to their historical asset data, firms can use AI to integrate real-time feeds and third-party data sources to create more accurate and dynamic asset valuations and forecasts.

    Foot traffic analysis is a current example. Previously, companies measuring where people spent the most time in a building had to rely on error-prone manual counts and statistical projections. More recently, systems that use cell tower pings appeared that provide data-driven proxy counts. However, this broad location analysis isn’t very useful for getting data about specific buildings as it only works down to a city or a neighborhood level.

    The advent of visually driven analytics opens access to continually updated location-specific data. Leveraging existing camera feeds, artificial intelligence can use object detection and person counting capabilities to deliver real-time shopper insights in and around buildings. AI can incorporate demographics, behavior patterns, and sales forecasts into traffic analysis. Asset managers can more accurately measure a building’s internal performance and compare it with similar assets in the same area or across an entire region. Before AI, it was too difficult and complex to incorporate all these performance data relationships into valuation models.

    Environmental, social, and governance data provides another example. Just a decade ago, ESG was talked about more than measured. It was nearly impossible to manually gather and analyze the vast amounts of data from utility bills, meter readouts, energy spend, and more – and not just for individual buildings. Asset managers would also have to gather these metrics across different portfolios and regions and account for environmental regulations in different parts of the world. Thanks to artificial intelligence, ESG data is now common in evaluating investment decisions.

    Continuing investment in AI development will help real estate asset managers incorporate more diverse data points that influence property values, including social media metrics. Asset valuation models could include the frequency of Instagram posts or TikTok tags related to a property, providing insights into public interest and buzz around a location. Asset managers could also value a location’s convenience and accessibility by the availability of immediate delivery services from platforms like Grubhub and Amazon. Similarly, artificial intelligence could account for negative factors like a lack of healthcare services or nearby grocery stores (aka food deserts). By integrating third-party data sources and real-time feeds from market aggregators into deep learning models, asset managers can deliver more comprehensive, realistic, and dynamic assessments and predictions of a property’s worth.

    It’s important to remember that these data points are only near-term ideas. As new data types and correlations are discovered, investment firms will need to update their asset valuation models to maintain a competitive advantage. As with the expansion of data, AI-powered tools will be required to facilitate the increased workload. Firms can avoid increased headcount costs while asset managers focus on their existing, value-creating tasks. 

    Advanced scenario modeling and forecasting

    Artificial intelligence doesn’t just make routine tasks more efficient. It also helps asset managers get more actionable insights and make smarter capital investment decisions.

    Scenario testing

    Asset managers deliver the most value though scenario modeling – testing the effects of different combinations of asset and market variables on portfolio assets and potential acquisitions to determine the most optimal investment decisions. As with any analytical model, better and more accurate data produces more actionable results. But many managers rely on static models that use general market variables that must be keyed in. This time-bound information constantly places asset managers behind the information curve.

    AI can enhance scenario modeling by continuously updating asset evaluation models with real-time data from both portfolio assets and market sources without human intervention. Manually fed point-in-time data can mislead static models, introducing not only wide variations in the data but also inaccuracies stemming from data entry errors. AI can continuously filter the data feeds, smoothing out trendlines, distinguishing anomalies from outliers, and flagging data points for further review to ensure data is being viewed – and used – correctly. What’s more, AI models are smart enough to know that today’s data doesn’t predict future performance. They can uncover trends in the data to provide guidance on future outcomes.

    Streamlining analysis with NLP

    Artificial intelligence will transform more than the way analytical models use data. It will change the way asset managers interact with data analysis itself.

    Major strides are being made in the user experience for testing scenarios. Using Natural Language Processing (NLP), asset managers can query complex market and asset scenarios with simple questions. It’s not just a chatbot performing a simple lookup of portfolio data and returning an answer. The asset manager could use a tool like MRI’s Ask Agora to enter a query like “Show me a model with interest rates at 5% versus at 4.5%.” AI becomes a smart assistant, aggregating data and performing complex calculations to deliver valuable insights.

    AI can also greatly expand an asset manager’s analytical capabilities, easily managing the complexities of testing multiple parameters. Rather than tweak individual sections in spreadsheets, an AI model could manipulate multiple variables simultaneously to provide a variety of outcome possibilities. It’s another way that artificial intelligence delivers actionable insights for asset managers along with time savings.

    Automated risk management

    Asset managers don’t create value just by uncovering new opportunities. They also do it by avoiding risk – and artificial intelligence helps them be more proactive and protect returns for investors.

    Anomaly detection

    The sooner asset managers can spot oddities in their data, the sooner they can address issues that affect portfolio value. AI models are excellent for detecting patterns across data wide sets, which will make them a critical component of risk identification. When looking at the rent payment history of all tenants across a portfolio, AI models could spot businesses consistently late in making payments across multiple regions. The report would alert the asset manager that the business is not doing well. The asset manager could then de-risk that holding before it causes a drag on returns.

    Market trend forecasting

    AI can also analyze macroeconomic factors like interest rates and other indicators to predict financial risks and opportunities. Consider the current high interest rate environment. As we’ve seen, AI models provide scenario testing across many more variables than humans can handle. The trend forecasting ability can show how these scenarios affect loan repayments for different deals across the portfolio. A falling interest rate trend could signal asset managers to attempt refinancing at a lower rate.

    Trend forecasting isn’t limited to financial data. Artificial intelligence models could discover correlations in historic weather patterns and foot traffic data provided by solutions like MRI OnLocation for Footfall Analytics. The results can show how assets whose performance is affected by weather (e.g., venues, outdoor malls) compare to similar assets that aren’t weather-dependent. Similarly, AI can analyze foot traffic data to compare an asset’s performance before and after remodeling to determine the ROI, and if similar remodels could improve performance in other locations.

    Workflow automation and asset management efficiency

    Every artificial intelligence discussion involves automation, but it helps asset managers specifically in more ways than you might think. By integrating multiple work capabilities, AI not only streamlines day-to-day operations but also enhances overall operational efficiency.

    Routine task automation

    First, there are the daily chores that come with the role. Task-driven AI systems can take over routine portfolio rebalancing, data loading and validation, and communications. For example, an AI assistant could proactively manage routine tasks by understanding deadlines and compliance requirements. The system could ensure timely submissions of legal documents or construction permits, thus avoiding potential fines or penalties.

    Enhancing operations

    As machine learning capabilities expand, investment firms will rely more on agentic AI. These goal-driven virtual assistants can use data about an asset manager’s calendar, communications, and workload to provide more proactive help. Instead of acting on requests for information or analysis, a connected agentic AI could automate processes related to task management and prioritization. It could book meetings, manage schedules, or prepare complex reports on the manager’s behalf. All actions would be based on the manger’s preferences and past actions, avoiding disruptions to existing workflows.

    Take an asset manager presenting at an annual client earnings conference. An agentic AI could use the meeting details and presentation information to not only create the necessary reports, but also finalize travel arrangements and book any necessary onsite conference rooms. The AI understands the end goal and the preferences and requirements involved, so all tasks can be created and completed autonomously.

    Compliance support

    Perhaps the most value from AI assistants is prioritizing tasks based on due dates and importance – especially concerning compliance with contract terms and regulations. AI-enabled systems will be able to analyze contracts, lease agreements, and other regulatory requirements. They will then create associated tasks lists to ensure timely completion of critical activities tied to deadlines, avoiding fines and penalties. For example, what if an asset manager had to provide energy consumption and carbon emission levels to a regional control board? Artificial intelligence could gather those readings from the building’s meters, prepare comparisons to similar assets in the area, and transmit the findings – all without manual intervention.

    AI could also ensure debt and lease obligation compliance. The digital assistants could monitor timelines found in contract clauses and alert managers about upcoming deadlines for debt management and lease agreements. These automated systems can help firms be more proactive when their team must act to align with contract requirements.

    Intelligent investor portals and relationship building

    Making asset managers more efficient isn’t just about AI taking on the manual data gathering and analysis. There’s also a sometimes overlooked, yet closely related, part of their jobs: client relationships. More specifically, client requests for information – and their potential to take time away from managing portfolios and evaluating deals.

    Maintaining good relationships is important for building investor confidence and engagement, and a key element in securing more capital. But client relationships also often involve time-intensive tasks like looking up very specific information or producing customized reports. With the growth of informed and data-hungry investors, asset managers see more of these questions every day. Responding takes focus off managing the portfolio, uncovering new deals, and creating value – not to mention missing information that could impact returns.

    Increasing self-service capabilities

    Most firms offer investors self-service portals, but they often only provide basic, high-level information summaries and reporting. They don’t give users individualized access to asset-level metrics or analytical capabilities that today’s data-hungry investors want. This is where AI can shine. Instead of a limited source of information, investor portals will be transformed into dynamic and interactive tools.

    The first step to AI-enabled investor self-service will be high-powered chatbots. Like customer service versions in the consumer market, investment firms can implement artificial intelligence trained on large language models to streamline responses to routine investor questions. Instead of taking time away from asset managers, AI can deliver fast responses to queries about current portfolio and individual account values, current fees and invoices, upcoming earnings reports, and more.

    However, major advances in natural language processing will be the cornerstone of investor self-service, combined with access to data analytics. As with asset managers, artificial intelligence can integrate diverse data sources, including commercial and residential real estate, asset management, and facilities data along with regional and economic data. While firms could put this information at investors’ disposal through the portal, not many users would have the analytical experience to make sense of the data.

    Instead, a user could ask the chatbot questions like “What are the most profitable locations of Company X in the portfolio?” AI would perform the requested analysis, and then use ChatGPT-like generative capabilities to send actionable insights and reports directly to the investor. Users can quickly obtain relevant answers and visualizations from vast amounts of data, freeing asset managers from performing hours of manual data manipulation and complex reporting.

    Fund reporting as a living document

    Speaking of reporting, AI has the potential to complement the firm’s financial reporting with customized, on-demand reports to individual investors. Traditionally, reports provided a snapshot view with limited metrics. Artificial intelligence will transform reports from a static record to a living document that investors can tailor to their needs. Expanding on the portal’s natural language processing and generative AI capabilities, investors can generate customized reports on finances and asset performance featuring the data points and metrics most relevant to their needs. Any number of reports can be run covering any number of analyses all without involving the asset manager.

    Going further, reporting will become like Microsoft Co-Pilot: a virtual assistant can examine an investor’s portfolio details and previous queries and then suggest different ways to build reports, as well as build and send them without the investor having to ask.

    Another benefit to AI-enabled reporting is that the reports will become more meaningful to investors, since the insights will be tailored to the metrics they care about the most. Meeting these investor needs more accurately – and quickly – will help boost investor confidence and retention.

    AI-generated benchmarks

    Every investor wants to stay informed – not just about individual asset performance, but also how the overall fund is performing relative to its peers. However, the amount of data needed to paint a full picture of assets (including facilities and maintenance costs, energy consumption, taxes, and other data points), the broader real estate market, and the economy are simply too much for the retail investor to take on alone.

    Only investment firms have the resources to compile that broad amount of data, giving them a way to give investors more visibility into performance and boost fund engagement. AI-driven benchmarks will provide valuable insights by showing investors their portfolio performance compared to similar assets across a broad client base. Instead supplying raw data, AI models will analyze industry-wide metrics like asset valuation growth within specific geographies or asset classes as well as performance across the breadth of the firm’s client base within the fund. For example, if similar assets are experiencing a 10% growth in valuation, these benchmarks show investors whether the performance of their holdings is more than general market trends.

    As these capabilities become more accurate, AI-driven benchmarks could also be used predictively. Based on historical data trends and current market conditions, investors (and asset managers) could get forward-looking perspectives on performance. Moving beyond basic metrics like interest rates or employment figures, these AI-driven benchmarks will correlate various factors to offer real-time, location-specific insights into asset and fund performance as market conditions evolve across multiple metrics.

    Benchmarking capabilities will not only help investors and asset managers make informed decisions about where to put capital. It will also help firms create a competitive advantage based on how well they curate the quality of their data sources and capabilities of their benchmarking technology.

    Using AI responsibly – and ethically

    One final, important note: While artificial intelligence will transform real estate investment, it’s still just a tool. And like any tool, AI can be prone to errors when not used responsibly. Ongoing issues with AI-generated images and text have shown how critical human responsibility and oversight are to AI-driven frameworks to prevent costly mistakes. It’s not just a matter of what’s legal; it’s also a matter of what’s ethical.

    All firms using AI must know best practices for data privacy and security as well as IP protection. AI tools don’t always surface their data sources, especially models trained on publicly available data. Without guidelines, AI outputs could comprise content from other companies – including competitors – and may create IP and security risks.

    The most successful organizations will establish use policies and ensure their AI-enabled solutions are trustworthy and accountable. For example, MRI Software uses a responsible AI solution development framework to avoid ethical dilemmas, bias, privacy breaches, inaccurate data, and liability. Data abstraction tools like MRI Contract Intelligence let users review, edit, and validate extracted data before using it in other systems. In addition to following data privacy and security laws, solutions don’t collect, use, or share data without consent.

    The right safeguards ensure AI models are reliable, fair, and transparent, with human confirmation of outputs. When firms are accountable for everything AI produces, their investors don’t have to be.