Newmark’s Wayne Gearey: 5 Questions

Commercial real estate services giant Newmark announced Tuesday that Wayne Gearey has been appointed head of geoanalytics and econometrics at the Manhattan-based firm.

Last week, PropTech Insider discussed Gearey’s role and vision for integrating geospatial artificial intelligence (geo AI) into CRE. With a background in geographic information systems (GIS) and machine learning, Gearey plans to combine GIS, APIs and geo AI to enhance location selection — which he calls location economics — for clients.

Gearey plans to bring geo AI to Newmark by combining AI-assisted data scraping with geographic and economic modeling to forecast disruptions before they occur. It should also be able to pinpoint untapped market opportunities and future growth zones, as well as equip occupiers and investors with an edge in timing, positioning and profitability.

The following conversation has been edited for length and clarity.

PropTech Insider: You joined Newmark in late July. Why did you choose Newmark, and how are you bringing location economics to the firm?

Wayne Gearey: One of the reasons I came here was the people. The leadership is fantastic. Specifically, Liz Hart [president of leasing for North America] on the business side, and on the IT side, Sridhar Potineni [chief information officer], who I’ve known for several years. I really enjoy working with those two people.

We plan to make it practical for brokers by providing modeling and having a technology platform with the ability to do that, where you’re fully aware of the capabilities of geographical information systems and you know what location economics is. There’s a foundation there for theory and methodology that would be built in immediately. 

On the econometric side, you’re essentially combining workflows. AI is really new to the process, but GIS and the application programming interface (API) for data, bringing in multiple data sets, is not. However, bringing geo on top of the AI — so geo AI — makes a big difference in how we’re going to help our clients get to the right location with confidence.

There are proptech companies addressing location technology. What differentiates what you are doing, and how does it help brokers?

Quite frankly, when you think about AI, the skills needed are very interdisciplinary. So what got me there is working on the front end with a lot of really interesting brokers. I started my career with Roger Staubach, being head of research there. When you go out on pitches, you can hear the questions being asked. So, when it comes to machine learning, you can start to build in a lot of those common questions that have been asked by clients and brokers. This certainly improves the work plan when we put all these in. So I have already started creating a workflow that combines those key entities.

The secret sauce is going to be the type of modeling that will be done — the predictive modeling. Mostly what you had before was very descriptive in nature, so people would respond to data that wasn’t indexed. With AI, you can teach your machine and use AI components to be much more predictive about a location. You can do scenario simulation much easier than you ever have before. Scenario simulation hasn’t been done a lot in the past. But now, not only can you do it for one location, you can do it for many locations at the same time.

How do you define geo AI?

If you tested ChatGPT, for example, you would see that it actually has a really good understanding of geography. If you said, “What neighborhoods exist in New York?,” it’ll come up with an answer. It may not be 100 percent correct, but it has an idea of boundary systems. Boundary systems are really what we use to make decisions. The lower the system, the better. So a neighborhood-level understanding makes looking at any issue, like crime, much more accurate than looking at it at a state level. 

On top of that, those boundary systems change geographically around the world. We have created a technology that connects those boundary systems together. We have the ability now to do global analytics on the same plane and the same scale, versus different systems trying to talk to each other,

How do you accumulate and clean this huge amount of data?

There are companies that have helped us out with that. We are collecting our own data and creating our own data, but there are companies like Placer.ai, and lots of other data companies like Chmura or Lightcast for labor, and ESRI, who are connected into almost every government around the world. We’re able to take their data and bring that into a data lake that is filled with sources, APIs and all kinds of things where we can create tools that will help us make decisions much more rapidly than before.

You have said that the improved accuracy and predictability of geo AI over traditional GIS is particularly effective in managing risks and supporting decision-making across various real estate sectors. How so?

Looking at all the things that we can do, risk prevention is definitely one of them. The Internet of Things is huge, and, with it, AI and geo AI. 

A good example is flooding. We have put things to measure flooding into most of the rivers across the United States. We can tap into that line. We also have indicators for where power outages are frequent. We know from satellite imagery where every single fire going on right now can be seen around the world. And, so, when you start to have access to all that data, you can start to future-proof risk a little more and have clients be ready to respond or understand what the risk is going into a location.

Philip Russo can be reached at prusso@commercialobserver.com.