Archer Uses Data Sets to Expand into Underwriting Tools

Real estate investment firm Archer has expanded its range of tools to help investor clients. The latest announcement is called AIM Automated Underwriting. The data-driven system uses machine learning technology to reduce the time needed for a first underwriting of a multifamily property to less than 15 minutes, according to the company’s claims.

“What we’re doing at Archer is help our clients find off-market deals,” Fred Canney, CFO and COO, tells GlobeSt.com. “We built the data platform to handle any commercial property, but we’re first focused on multifamily. We’re used machine learning and put in deep-learning models to find the off-market opportunities. What we’re hearing from clients is that their teams are stretched. We see repeatedly that it takes time for our clients to underwrite things and evaluate opportunities.”

One of the biggest time wastes in commercial underwriting, according to Archer and other companies developing similar offerings, is data entry.

“The underwriters I’m talking about are multi-family equity acquisition folks,” Canney says. “These are the acquisitions analysts and associates. We looked at the repeated activities and data. You can preload an acquisition model with all the data someone needs to make a good decision.”

For example, someone could pull up data like rent forecasts from different sources and choose one or pick a comp model out of multiple alternatives. “We’re trying to arm these folks stretched for time trying to do more with less—to do the grunt work [for them]—and they can choose what to use and the decisions they want to make before and investment,” says Canney.

The pattern follows one becoming prevalent in proptech, in which software is supposed to undertake repetitive processing that had, out of habit and lack of alternatives, been relegated to underwriters and other professionals.

Archer has been expanding from its foundational set of data with software to take on different tasks. “We started with helping them locate properties to acquire,” Canney says, addressing questions of supporting decisions about capital sources, investment levels, and strategy. The company has a location recommendation engine in which a client enters a location and a type of property, like a class B building that needs renovation, and use various types of data to identify others that appear as though they’ll be on the market before being listed.

The biggest advantage of these automations—generally depending on using the same data in different ways, is to cut the time needed to complete work, giving investors a first mover advantage.