Applying Spatial Intelligence to CRE Operations

Erik Sherman

Counting crowds has become an important tool in retail: that is, watching where people go in a store to better understand how they look at products and make purchase decisions.

It’s a lesson that commercial real estate should learn, argues George Shaw, CEO of Pathr, a software vendor that provides tools to study the movement of people in buildings without collecting identifying information, to better preserve privacy. Originally designed for the retail industry, Pathr is expanding to other industries, including commercial real estate.

“Thinking about office buildings, where we’re doing work with an operator of quite a few office buildings worldwide, they wanted to understand how long someone is willing to wait for an elevator,” Shaw tells “How long am I going to wait tomorrow if I arrive at 9am versus 10am? How long will I wait for a treadmill at the gym? How long will I wait for coffee? When a tenant shows up, are they greeted by the guard?”

The data provides information that owners and operators can use to make better decisions. For example, by watching where people go in an office building and when, there might be patterns of times where more people congregate. That would offer insight into how to distribute HVAC services, increasing heat or air conditioning in more crowded areas and reducing it where lower amounts of power can handle the number of people there.

Other examples: Monitoring elevator use and patterns could help plan when to clean nearby bathrooms, refill hand sanitizer dispensers, shut down some elevators during slow periods to save energy, or arrange a casual event for tenants, like coffee.

“Even thinking about people walking and foot traffic, there are a lot of interesting changes that come about,” says Shaw. “That also applies to digital signage and operation in a mall. How often you empty a trash can depends on the number of people walking by the trash can. How often do you clean a lobby? Understanding the foot traffic has a huge impact on how to operate a building.”

Having the data is one step. Optimization can be another, requiring expertise and techniques like machine learning.

“It implies us being diligent and honest as well with our customers about what the data’s saying and what it’s not saying,” Shaw adds. “Let’s talk about the optimization problem you want to solve. If it’s that every person should be at 72 degrees and minimize the amount of energy used, that’s a problem data science can look at.”