It’s the worker that never sleeps. The real estate industry is embracing machine learning technology as new tools take on mundane administrative tasks, such as powering advanced accounting systems, automating marketing campaigns, and creating smart contracts.
But back-office efficiencies are only one component of the advantages AI and machine learning bring to the table for real estate professionals and REITS. Where machine learning really comes into its own is as a tool for predictive modeling for property analysis and investments.
According to the Center for Real Estate Technology & Innovation (CRETI), in the first two quarters of 2022, $13.1 billion was invested in PropTech companies, companies that provide data and analytics to property investors. That $13.1 billion marked a 5.6% year-over-year increase over 2021.
Proptechs provide technology tools so real estate professionals can effectively buy, sell, research, market, and manage their properties and portfolios. Read on to find out why the PropTech industry is booming, and REITS are targeting quality investments using big data, AI, and machine learning.
Why REITS Are All About Data and Machine Learning
According to a survey by Deloitte, real estate investment trusts (REITs) are taking full advantage of artificial intelligence and machine learning and “leading the AI revolution.” In contrast to executives in many industries who have not yet fathomed the possibilities of AI technology, forty-one percent of REITS’ executives are fully on board with algorithms and machine learning models.
Where AI is faultless is its role in assuming rote administrative tasks, and there are plenty of those in property management, such as sending scheduled marketing emails or sending out automated rent payment reminders. But for REITs, AI and machine learning are affecting much more than the day-to-day administration. These technologies are shaping long-term investment strategies.
Here are some applications of big data, artificial intelligence, and machine learning that are disrupting the REIT sector.
Financial Modeling and Property Selection
Machine learning investment models analyze all sorts of data in the REIT space to create investment strategies. REIT financial modeling is based on accurately estimating the present value of a future source of cash revenue a property will generate. These financial models are based on historical data, and many unknown variables that will change the markets and economic climate in the future. Thus, machine learning models are not the only driver of investment decisions but are used as screening tools to help analysts improve stock selection and investment opportunities.
With machine learning solutions, creating a pro forma for any apartment building or property anywhere in the country simply by putting in an address and generating potential returns is possible. Property investors can compare comps and make better and faster decisions.
Tracking Performance Relative to Competitive Sets and Equities
Property investors already use AI and machine learning to create financial pro forma for properties and commercial real estate. Near real-time revenues can be pulled from market data services or community websites using data scraping technology. Expenses are estimated using existing portfolios or data from digital data providers. Machine learning algorithms can then identify competitive property sets and determine if a real estate asset is underperforming or outperforming the competitive sets in real-time.
Publicly traded REITs are often more closely correlated with equities rather than real estate assets. Analysts can apply data and machine learning to track the value of the underlying assets relative to the stock price and inform the analyst whether a public REIT is a buy or a sell.
In early 2020, the Federal Housing Finance Agency allowed remote appraisals for purchase loans to become permanent. The limitations placed on appraisers during the COVID lockdown paved the way for automated valuation models (AVMs) and desktop appraisals drawn from public records like tax appraisals, listings, and other digitized property information.
Appraisers used to rely on cap rates for the trailing 6 to 12 months. The cap rate is calculated by dividing a property’s net operating income by its asset value. Unfortunately, pricing is often set several months prior to a closing when the buyer is selected after bidding. Using historical data that goes back even further, say eighteen months, investment firms can track and analyze the sales data and gain an advantage in the future by seeing earlier movements in the markets possibly 18 months prior to the traditional appraisal industry.
AI-based appraisal solutions are also highly sophisticated. They incorporate aerial imagery, machine learning algorithms, and computer vision to gather property condition data on properties nationwide. Details such as the type and age of roof construction, the presence of solar panels, the size of the property, and whether the property has a pool and a tamed yard can all be gathered without needing an in-person visit.
Automated appraisals also eliminate subjectivity and human bias. Technology can provide a more objective valuation for real estate investors through machine learning and big data. Automated appraisals reduce bias, speed up turnaround times and deliver quality, on-demand appraisals at scale for more profitable real estate decisions.
The best business decisions are data-driven. But big data and machine learning allow analysis on a micro level to be combined with macro-level analysis and human expertise. This type of analysis avoids generalizations and can factor in data that impacts real estate investment more precisely, such as climate, noise pollution, and competitor activity.
McKinsey & Company expected insurance-based underwriting to become automated in the future to gauge risk more accurately, eliminate bias, and provide standardized premium payment rates. A machine learning model can use historical data to translate the risk factors into a suggested premium. The model can determine if there are any changes to the most important risk factors when renewing contracts and automatically generate quotes based on the previous year.
Investor and Client Relationships
It may seem impersonal, but automation enables better communication among investors, clients, and other stakeholders. Apps and chatbots provide 24/7 attention to queries and keep everyone updated on the progress of projects. Mobile apps give investors the power to manage portfolios wherever they are, drive successful strategic operations, and improve asset management.
ATTOM Is Front and Center of the PropTech Movement
The decisions resulting from machine learning models are only as reliable as the data used for those models. Property data providers, like ATTOM, are constantly updating their datasets and collecting new data points that real estate investors and professionals can apply in their strategies.
Examples of our data include big-picture elements, such as flood, drought, and environmental hazard risks to property level valuation models like AVM and rental AVM, which predict future income from properties and rental properties. We also provide neighborhood data, such as sales, local points of interest, schools, employment, and crime data.
Failing to invest in technology is short-sighted. Unlock your potential with ATTOM.