How AI and Machine Learning is Impacting the Real Estate by Roy Dekel

Artificial intelligence has become the breakout technology in the past ten years, utilizing huge amounts of computing power to learn and identify patterns in data without the guidance of humans. These algorithms can be used on nearly any problem or question, provided there is enough input data for the algorithm to process to generate realistic results. This broad generalizability means that industries that have traditionally relied on purely human-driven research and development can now harness massive amounts of data to become more efficient – and potentially more profitable.

A sector undergoing this technological shift in recent years is the real estate industry. Since its inception, property ownership has been driven by personal preferences and human interactions, primarily among buyers, sellers, and real estate agents. But now, with algorithms able to process individual preferences and generate data at a room-by-room level, massive changes to the buyer-agent-seller paradigm are happening.

One man who has been extremely influential in the real estate industry’s technological revolution is SetSchedule’s CEO and founder Roy Dekel. SetSchedule has developed a multi-patented agnostic leads marketplace featuring leads from top tier real estate publishers and lead generation websites. The marketplace alongside a multi SAAS products suite, leverages AI-powered predictive analytics, big data and machine learning technologies to deliver better business efficiency to agents, teams, brokers and industry verticals. Additionally, before founding SetSchedule, Roy led a high-profile investment fund that bought houses the traditional way.  

During his time as leader of that investment fund, Roy learned some valuable lessons on the efficiency of the traditional real estate market. According to Roy, “curating leads, identifying customers, rigorous follow-up,” and a plethora of other tasks related to linking potential buyers with current sellers was “extremely daunting.” The enormous amount of energy and time spent on a single property sale seemed quite inefficient.

With this in mind, Roy identified some key areas where technology could influence the space – and in particular, areas where AI and machine learning could significantly disrupt tradition. One key area where AI is already influencing the market is in home value appraisal (also referred to as comparative market analysis, or CMA). When a person approaches a real estate broker intending to sell a house, the broker will analyze key data points to appraise the potential selling price of their home. Factors like location relative to schools, shops, and other homes are considered, as well as square footage, in-home upgrades, and more.

Unfortunately, data analysts can only process a few factors in a realistic amount of time. But machine learning algorithms can intake huge amounts of data, process all of it to find patterns in home value, then return highly accurate valuations for homes in a relatively short period. Factors like neighborhood volume, the average number of cars passing through, and even vegetation on lawns can be used in these algorithms. Roy thinks that this is one of the key uses for machine learning, stating that “machine learning software can analyze all the factors…significantly increasing the accuracy of an agent’s analytics”. Rather than replacing agents, these tools can be used to augment the information an agent has during the sale process.

Another aspect of the real estate industry that’s using AI is the targeted advertising space. While targeted advertisements have been part of Google and Facebook’s wheelhouse for years, other industries have found it difficult to link buyers and sellers without hands-on time from experts in the space. In real estate, this is especially true – the exhaustive lead generation and customer vetting that Roy referred to is the industry standard for advertising.

Fortunately, recognizing patterns in individual-level data to find potential homebuyers is a fantastic use of deep learning algorithms. These algorithms can be used in places where homebuyers are likely to search for homes (and in some cases, on social media platforms like Twitter and Facebook), scanning for buyer profiles that match homes on the market. By identifying these eager buyers, algorithms can highlight potential leads that are more likely to be interested in the homes that agents are offering.

This is the principle behind SetSchedule’s Smartmatch software, which scans through all of the data generated by thousands of agents to understand which leads resulted in successful home purchases. Other services like Coldwell Banker’s CBx app are also starting to use this technique, signaling a massive shift in the way that agents and homebuyers seek each other out for upcoming deals. Roy thinks that these artificial intelligence-driven apps will be critical in “assisting brokers, teams, and individual agents to find clients,” but won’t replace the human touch of the real estate industry.

Technology is critical to most aspects of everyday life, and industries that rely on human interactions still are feeling the impacts of technology. But the applications of algorithms used in machine learning can be viewed as a tool for sectors like real estate, as Roy Dekel suggests. By harnessing the power of machine learning, real estate agents can increase the efficiency and productivity of their sales pipelines while giving homebuyers the home of their dreams.