Data powers AI solutions, but developing the capabilities to gather, structure, and deploy the right data is a daunting challenge. For the real estate industry, the task becomes a lot easier when leaders know which questions they are trying to answer.
Katy McLaughlin: Let’s say a real estate leader wants to take advantage of some of the new AI tools. How do they approach working with data to get started?
Jules Barker: They don’t. Before they get anywhere near data gathering, structuring, or feeding data into any kind of model, they need to ask themselves what they want to do first. They need to step back and ask, “What are we struggling with? What’s our competitive advantage, or what should it be?”
Then they need to understand the process and talent challenges these problems pose. We find that most organizations have great ideas but struggle to execute them. So they need to think through what data and system they need and who can help deliver it.
Katy McLaughlin: Can you provide an example of a problem asset managers face that can be answered with an AI-based solution?
Jules Barker: Let’s say that you’re an asset manager whose teams are bogged down by hunting for pieces of information about properties, leases, and other asset information. Multiple programs and systems are slowing down your teams and creating risk in the form of errors, multiple checks, and rework. You want to make your people more efficient and effective by providing them with a single source of truth that’s accurate and easily accessible.
The solution could start by using AI to read piles of paper leases, even those written in various kinds of script and with handwritten notes on them. Digitizing and structuring this data into a single, secure, and governed environment can often reveal all kinds of gremlins, such as contradictions about the exact square footage in a property or the exact day a lease expires. Once those discrepancies come to light, they can be resolved.
Correct and coherent data will allow a company to move faster and more confidently. Instead of spending hours on data gathering and validation, people can reinvest their time into making quicker and better decisions, which can ultimately enhance company performance.
Katy McLaughlin: What’s an example of a problem a building operator could address with better or more organized data?
Jules Barker: Let’s say the question is, “How do we reduce energy consumption across the portfolio in a way that creates positive experiences for our tenants?”
Historically, there would be one energy feed into a building and a bill at the end of each month. The property manager would just attribute a bill to tenants based on their square footage. But today, smart meters reveal who is using how much energy at which time of day. We’ve observed that just by seeing their own specific energy usage, tenants are equipped and incentivized to reduce consumption, in some cases by 30 or 40 percent. Especially when a building’s system data is integrated into a single dashboard, it’s easy to spot inefficiencies.
Katy McLaughlin: What kind of data are buildings able to collect today that they couldn’t, say, five years ago?
Jules Barker: In the office environment, until relatively recently, it was pretty expensive to do anything more than count the number of people swiping into the office every day. Over the last five years or so, it’s become almost ubiquitous for offices to create a more granular view of how space is used through footfall counters, Wi-Fi systems, passive infrared sensors in the ceiling, and Bluetooth beacons.
But the real innovation today is not a particular single information feed; it’s the ability to bring data together in one place. Granular internal data can be layered with, say, historical property performance data and local and regional demographic data to create a comprehensive view.
Katy McLaughlin: What are examples of questions that real estate companies can answer with both a granular and comprehensive view of data?
Jules Barker: Take a market-savvy investor, for example, who also understands in absolute detail what it will take to bring a building up to the required sustainability standards. They can bid on it faster, or bid more for it, and capture an advantage by doing so.
Or take a mall operator who really understands local demographics and the detailed footfall dynamics through the mall. The operator can more accurately target the best possible tenant for a particular retail outlet.
Granular data about tenants’ behavior and space usage can be really helpful for churn analytics. It can be layered with nontraditional information—including proximity to public transportation, quality of nearby restaurants, and local footfall—that’s informative about likely occupier behavior. The right data can enable an owner to predict which tenants are more likely to move out when their lease ends, which can inform how to work with tenants to improve retention.
Katy McLaughlin: How do real estate companies figure out what problems to tackle first?
Jules Barker: That’s the most important question! It can be tempting to gather and analyze as much data as possible to work out the biggest challenges and opportunities. But we strongly believe that companies shouldn’t just take all their asset, operational, and market data and shove it in a big pile, even if they’re then going to structure all of that. This can add years to the process before generating any results. It can also be counterproductive, because companies could find they’ve gathered the wrong data in the wrong way for the problems they later realize they want to solve.
Instead, teams within real estate companies need to break down the challenges they’re facing and the upside opportunities they want to capture with AI and gen AI. Then senior leadership can prioritize projects based on estimates of how long each one will take, the ease of implementation, and the potential for impact. We’ve found that companies can nearly always identify the quick wins and highest-priority transformative use cases. Then they can get started on no-regrets actions that rapidly yield benefits. Importantly, this can demonstrate the momentum to stakeholders who might otherwise be blockers to change.
Katy McLaughlin: Any last considerations that can lead to wins?
Jules Barker: Any tech build requires continual design and scoping choices that, cumulatively, shape the end product. Combined business and tech teams with an active C-suite, held accountable by the CEO as sponsor, are critical for success.
