For at least a decade, big data has gone from buzz word to everyday vernacular. But within the commercial real estate industry, claims of big data usage have been a marketing gimmick at best and downright fraudulent at worst.
This isn’t because of the lack of data to work with. The sheer quantity of data commercial property owners and operators have at their fingertips is growing daily. The problem is that only a small subset of this data is utilized. It is no exaggeration to say that the industry has near-endless possibilities for unconnected datapoints. Many of which are often ignored entirely when they could be used to support tenant satisfaction and bottom-line performance. This black hole of disregarded data is something we like to call the “data abyss.”
In total, this abyss accounts for a staggering 97 percent of relevant data, meaning that property owners and operators are only using 3 percent of the information available to them.
This universally neglected data can be found in a host of siloed locations, including internal platforms such as Yardi, MRI, VTS, Salesforce and Microsoft Dynamics. Market and industry research are also relevant sources, providing local real estate data such as vacancies and closed leases, weather trends, and information about tenants’ business sectors. Other data sources include tenant experience information (from apps like HqO and Equiem) and data from HVAC or other building functions. Ideally, for the future of commercial real estate, landlords would find a way to leverage these sources, and proactively capitalize on the data to generate intelligence which could allow them to raise revenues and reduce expenses; however, tackling the data abyss is a challenge too vast for any one individual.
Enter artificial intelligence.
Part of the power of AI comes from its ability to detect what the human eye misses. AI leverages machine learning algorithms to catch complex patterns, nuances and anomalies that would likely be overlooked through human analysis. You or I might be good at sifting through an organized spreadsheet and identifying a pattern, but AI can examine massive amounts of data from a dozen different sources and understand how patterns and anomalies in one stream can impact on the building or portfolio more broadly.
Used as a tool for analyzing these troves of data, the potential of machine learning is almost limitless.
For example, at multifamily properties, a machine learning engine can look at both property and market data, and analyze the various characteristics that affect local rental rates. This helps leasing professionals pinpoint their strategy for leasing a vacant apartment: Based on demand in the market and unique characteristics within the building, it might suggest that pricing apartments at a lower rate, which would increase occupancy, will increase the building’s overall revenue. In some cases, it might determine, based on local data, that it is advisable not to put the apartment on the market immediately, as it would make more financial sense to remodel the kitchen and only then seek to lease the renovated apartment.
With regard to tenant comfort, an AI tool might observe that in specific weather conditions, tenants with a certain profile are likely to experience discomfort. Instead of waiting until tenant complaints come in, the platform would flag the likely impending issue to the appropriate property manager so s/he could take action. Similarly, it might identify that elevator issues tend to occur when cold weather is sustained for several days, and alert an engineer to take preventive action when a cold front is approaching.
In all of these examples, the value AI brings to the table is significant, because it successfully joins data from two or more independent locations to identify a trend; while humans (and their Excel files) are capable of flagging trends within a single dataset, their ability to glean insight from multiple unconnected data sources is much more limited.
In addition to the value of AI in bridging the gap between data silos, it is also helpful in that it enables individuals to deal with the amount of data that exists within the abyss. The abundance of information from all of the relevant sources is so vast that it would take years to analyze without the benefit of AI, years later, most of the data is no longer timely. Instead, property managers neglect the vast majority of relevant data because of the time it would take to analyze it.
By deploying machine learning that analyzes data from existing systems in real time, property owners would be able to generate insights in a timely manner, enabling building professionals to move quickly as they capitalize on these opportunities.
Realistically, the vast majority of CRE professionals are not data scientists; yet, the insights a data scientist might generate could help them meaningfully increase NOI. AI steps into the breach, supporting asset managers, property managers and other building professionals by alerting them to both opportunities and issues that would otherwise go overlooked.
AI is a powerful tool for CRE professionals because the insights it provides can have a positive impact on the property team’s daily workflow as well as portfolio NOI. And, as the potential for AI is growing increasingly clear, companies like mine, Okapi, are looking specifically to unleash AI across CRE portfolios.
It should be stressed that, in 2019, AI’s ability to enhance real estate operations is no longer a futuristic idea. It’s a tried-and-true reality. Okapi has worked with a number of the largest property owners in North America, and has, without fail, generated predictive insights to help boost revenue and leasing velocity and cut expenses. As a rule, we’ve been able to help property owners increase portfolio NOI by 1-3 percent, which, for a large portfolio, can translate into hundreds of millions of dollars in asset value.
By utilizing our platform, professionals at some of the largest CRE companies have gained increased understanding into how their assets are performing or underperforming in real time, and how they might perform in the future. Because our platform uses AI to gain predictive insights, it goes well beyond simply informing about past failures and successes, and provides suggestions to building and leasing teams based upon these insights.
Ultimately, this predictive ability can be used to increase tenant comfort and cut expenses, all of which contribute to raising NOI.
It’s no secret that the real estate industry was initially slow to embrace emerging technologies, but that it has picked up significant steam in the past few years. But while real progress has been made, there have only been a handful of companies that have truly tapped the power of machine learning to increase property operations.
As the cycle progresses and the industry becomes increasingly competitive, we can expect a much higher level of sophistication from property owners and managers. This will mean that we have to be honest when we throw around terms like big data. The truth is that right now, data in buildings is small. But this is changing quickly. One day soon we will be able to usher in the true era of big data in the property industry and make true on our promises.
Credit: Google News