Improving real estate decisions

With machine learning

By Matthias Muench on Mar 12, 2021 in Global

The human brain is capable of tremendous achievements. But what are its limitations in business transactions, specifically those involving property and real estate investment management? At what point do machine data-based systems make more accurate decisions than intuition?

Human intuition certainly has its place. As Deloitte researchers Surabhi Kejriwal and Saurabh Mahajan have noted, “The [real estate investment and management] industry has long thrived on relationships, which is how many investors have traditionally gained access to unique information. Traditionally, most investors have combined this information with their gut instincts to make investment decisions.”

But although intuition can be a useful tool, Harvard Business School Online writer Tim Stobierski cautions that “it would be a mistake to base all decisions around a mere gut feeling. While intuition can provide a hunch or spark that starts you down a particular path, it’s through data that you verify, understand, and quantify.”

A team of McKinsey experts echoes this sentiment, noting that complex decision-making requires analysts to “sift through tens of millions of records or data points to discern clear patterns and place their bets with few supporting tools to help glean insights from that material.” By the time the data needed to determine a course of action is collected, compiled and processed, they note, “the best opportunities are gone.” There’s also the problem of “cognitive biases” that misguide decisions with information drawn from the wrong sources.

Fortunately, Stobierski notes, “it’s never been easier for businesses of all sizes to collect, analyze, and interpret data into real, actionable insights” into portfolio health measurements such as revenues, debt, risk, occupancy and sales, along with property-level operations like energy consumption and accounts receivable. Ronald D. Marten, CCIM, writing in Forbes, adds that “CRE brokers who can tap into today’s sophisticated data tools can differentiate themselves and their core value proposition to clients. Knowing everything about a building by using flood maps, demographics reports, traffic counts, tenants and retailers … and more gives a potential buyer an accurate idea of what their ROI is going to be on day one.”

What do machine learning algorithms in the real estate realm consist of? One example is combined macro and local forecasts that identify areas with the highest demand for residential housing. On another front, retail mall investors can combine operational data at the property level with sales data from mobile sensors, social media and physical store sales, then use machine learning algorithms to analyze consumer buying behavior. Similarly, commercial property tenants can compare rent rates across various markets to make more informed decisions and get into spaces faster.

Data compiled from multiple disparate systems is complicated and prone to error. As a result, sophisticated software applications capable of collecting, processing and using data across the asset management lifecycle have been developed and brought to market. This technology, complemented by machine learning recommended actions, enable management of deals, budgeting, investor reporting and more in a single connected system.

Developers seeking new parcels, for example, can use advanced analytics to assess the properties’ potential, property uses and even pricing, among other things. Asset managers can evaluate pipelines and match deals with investors, benchmark their properties’ rent against others in the area, tie capital calls to investment lifecycle data and generate reports. Property-level data collected within a centralized location enables everything from online tenant payments to reduced heating, cooling and ventilation costs and better oversight of construction projects.

Kejriwal and Mahajan point out that “investors and managers can leverage analytics and AI across key steps in the investment life cycle, from deal sourcing to portfolio management to risk management. In addition, these technologies can help increase efficiency and effectiveness of operational processes, such as information integration, investment accounting, and reporting.”

Real estate software technology holds massive potential to shift decisions from humans to machines. Assimilating all asset management information at the property and portfolio levels and makes it universally available can preempt cognitive biases by shifting the decision-making onus from humans to data-based systems.

Advanced analytics can help investors and managers understand risk, from the asset level all the way through regulatory issues, while reducing human bias. They can produce faster decision-making, lower transaction and operational costs, and more rigorous risk management and portfolio optimization. Ultimately, Kejriwal and Mahajan say, “investors and investment managers can use data analytics and artificial intelligence in their existing acquisition, disposition, and portfolio management processes to manage rising risks and complexities more effectively.”