In the latest recognition of the impact of artificial intelligence on real estate, on Tuesday venture capital firm Thomvest Ventures released its first-ever Top 50 AI Startups in Real Estate list.
The list is composed of unranked proptech startups that have demonstrated a foundational use of AI in their operations and services for clients, said Nima Wedlake, managing director at Thomvest, a San Francisco-based group of investment companies that makes investments on behalf of billionaire Peter J. Thomson and his family.
SEE ALSO: All in the Family: Matias and Eglae Recchia On AI as a Commercial Real Estate Disruptor
In assembling the list, which the firm plans to do annually, Thomvest first gathered and analyzed data on AI in real estate in-house. It then presented its findings to traditional real estate companies, other investors, and non-AI proptech firms, asking what listed companies ranked high with them as AI businesses.
“We had the insights for the real estate AI universe that included around 750 to 800 companies, and the first filter for us was, ‘Is this AI that’s core to the product offering, or is it additive?’ ” Wedlake said. “There’s a set of incumbents out there that have embedded AI into their existing products, but that wasn’t the purpose of this piece.”
The percentage of proptech companies using AI is growing in a meaningful way, Wedlake said.
“It’s not just ‘Let’s slap on AI and that’ll double our valuation,’” he said. “It’s ‘This is changing the end product we can deliver to our customers or the use cases we can solve for.’ And, so, it feels meaningful and real. We thought this would be a good opportunity to aggregate all the activity happening at the intersection of AI and real estate, and then highlight what we and others in the investing community consider to be some of the more promising folks out there.
“This list will change. It’s a snapshot in time of who folks consider promising. When we do this next year, it’ll be a very different set of companies. I didn’t want to create just a market map that had hundreds of companies on it. I wanted to be a little bit more focused.”
Thomvest’s methodology for the list also included filtering companies based on geography, focusing primarily on North American startups. Also, the list excluded public companies that Thomvest considered too late-stage to be comparable, as well as some early-stage startups that lacked paying customers and general availability to the public, said Wedlake.
“Once we whittled that down to several hundred companies, we went out to most of the major proptech investors, a lot of the large owners of real estate that are the end users and paying customers of these tools, and said, ‘Within your portfolio, what are the two to three most promising AI companies? What are you piloting at the property that you find really interesting.’”
Among proptech startups that are actually AI-centric, Thomvest found pre-construction and construction tech startups to be over-represented on its initial list, Wedlake said, adding that a lot of AI startups are focused on selling into real estate trades verticals. The depth of market penetration for all varies.
“If you think who here is farthest along, I would say EliseAI, which is a unicorn now — they’ve been doing AI before it was cool,” said Wedlake. “I talk about the importance of creating immediate value for end users, because AI is sort of this shiny object right now. And a lot of people believe it will be important, but don’t necessarily know how to implement it in their day-to-day operations. I think Elise has been very focused on automating leasing workflows. There’s a clear [return on investment] story they can tell, and then they’ve done a great job commercializing it. They have broad adoption across a lot of resi-focused real estate owners. Kudos to them for having the foresight to build this even before Chat GPT launched.”
In addition, there are a number of AI-native companies that are building workflow automation for plumbing companies or for HVAC companies, Wedlake added. “I consider that proptech because they live in the built world, and they’re touching real estate. So we included a handful of companies that are selling into specific trades.”
Relying on the real estate incumbents, other investors and non-AI proptech companies’ input helped Thomvest create an agnostic list, said Wedlake. The list contains only two Thomvest portfolio companies: Keyway, a machine-learning technology platform for analysts, brokers, sellers and buyers to simplify transactions below $20 million; and mortgage technology startup Maxwell.
Thomvest has been investing in proptech startups for about seven years and now counts about 25 percent of its investments in that category as AI-centric, said Wedlake. He expects that percentage to increase.
“We’re investing in vertical software, and you have to be obsessed with the value that the product is creating for the end user, because revenue scales with value created,” Wedlake said. “The story of enterprise software is that you build a specific solution for a specific customer set — for example, mortgage originators — and that solution is better than any horizontal software they can get off the shelf. You look to win them as a customer and to get them to implement your tool. Over the course of the next several years, you build additional solutions that wrap around that existing wedge solution you sold, and then you grow with the company.”
As to which leads in today’s real estate marketplace — AI tech or the ability to market it — Wedlake hedged.
“I would say sales and marketing still matters meaningfully,” he said. “Maybe it’s old fashioned, but it’s just a fundamental truth to selling technology. At the end of the day, I would say the challenge for folks in real estate now is that there are more tools. There are more things you can do. The surface area of what you can do with technology has increased, and so it’s also easier to build software.
“Now, not only is AI impacting real estate, it’s also impacting software development. So to get to a product for end users, the time has shortened. As the product piece commoditizes slightly, the sales and marketing piece, I think it is more important than ever for a lot of these companies.”
Philip Russo can be reached at prusso@commercialobserver.com.