Security Properties, a national real estate investment company based in Seattle has launched a proprietary underwriting application powered by artificial intelligence (AI) that was developed in-house. According to Security Properties, FirstPass, trained on decades of the firm's underwriting and operating data, compresses the initial financial analysis of a multifamily acquisition from four to five hours of analyst work to less than five minutes. The platform has been in use across the firm's acquisition pipeline for several months, including transactions under evaluation.
FirstPass takes operating statements received during due diligence, including rent rolls and financial statements, and uses a machine-learning algorithm to classify the data and enter it directly into Security Properties’ underwriting model. The application pulls current U.S. Treasury rates and layers in deal-specific parameters such as transaction date, financing structure, location, rehab velocity, and rent-growth assumptions. Then, it generates a forward-looking view of operating performance across different scenarios. The platform is continuously utilized throughout underwriting and due diligence, updating assumptions as new information surfaces.
According to Security Properties, the goal is not only to increase efficiency from a workforce perspective, but to increase the firm's deal sourcing velocity significantly.
“FirstPass changes the math on how many opportunities we can seriously evaluate,” said chief investment officer Mark Bates. “We can now run a rigorous first analysis on a deal the moment an offering memorandum hits our desk, before we ever enter due diligence. That means we turn over far more stones to find diamonds. We cast a wider net, and we only pursue the very best opportunities for our investors.”
While AI adoption continues to increase, many questions remain around trust. Rather than licensing a third-party product, Security Properties built FirstPass on its own operational data, gathered across 57 years of investing through every market cycle. In addition, every output is reviewed by its team of analysts and acquisitions professionals before any investment decision is made.
“The quality of the output is a direct result of the quality of the input,” said Éamon Finneran, senior director of operations and technology, who led the development of FirstPass. “We are feeding the model clean, actual operational data from our own portfolio history, not scraped market averages. And the machine never has the last word. Our analysts and acquisitions team review every result, so what you get is the efficiency of AI paired with the subject matter expertise of a trained human eye. That blend is where the real advantage lives.”