9 Startups Using Artificial Intelligence in Real Estate
Everyone knows that probably the smartest investment we could make today would be in cryptocurrency. Said no one ever. Another sure-fire way to lose money is to invest in over-the-counter stocks, something we’ve been preaching for years. Regular readers who have been paying attention know that the safest long-term bet is to find an ETF with low management fees that follows a major index. For tech enthusiasts like us, that’s PowerShares QQQ (NASDAQ:QQQ), which tracks the Nasdaq-100 Index. But you’re not reading Nanalyze just to buy and hold index funds, but to learn about emerging technologies that might be worth an investment. So let’s talk about artificial intelligence in real estate.
The hype train on AI left the station a while ago, with the most recent quarter showing that private AI startups raked in nearly $1.9 billion across 116 deals, according to PwC and CB Insights. That’s about a 30 percent jump from the previous quarter and represents a two-year quarterly high. Now, let’s talk real estate: In 2017, real estate accounted for about 13.4 percent, or $2.6 trillion, of U.S. GDP. That’s more than any other industry including manufacturing at $2.2 trillion. Real estate can cover, well, a lot of real estate, from commercial categories such as construction and rentals to residential housing. For example, real estate construction alone contributed more than $1 trillion to the economy last year, while apartmental rental properties are worth north of $1.4 trillion.
We could keep throwing numbers at you—it’s a bloodsport akin to Aztec baseball for our MBAs—but you probably get the idea. We have a multi-billion-dollar technology sector ready to disrupt a multi-trillion-dollar industry. But like other industries, such as long-haul trucking and warehouse logistics, real estate moguls are slow to adopt the latest and greatest technologies. We’ve come across nine startups that hope to change that by applying artificial intelligence in real estate.
The Sky’s the Limit
An Israeli company founded just last year but with offices in one of the world’s most expensive real estate markets, New York, Skyline AI just raised a $3 million Seed round in March from top VC firm Sequoia Capital. The startup claims its platform can tell real estate investors what properties offer the best return by ingesting tons of data from more than 130 sources, taking into account over 10,000 different attributes on every property, going back as much as 50 years on every multi-family property in the United States. This is what the Skyline platform says it can do:
The four founders have a track record of starting and selling AI companies, so we wouldn’t be surprised if the endgame is another big-time exit, maybe to a real estate player like Zillow (NASDAQ:Z), which itself uses machine learning to put a price tag on more than 110 million homes in the United States with a reputed accuracy of 5 percent. Zillow recently announced it would get into the business of buying and flipping houses.
Update 8/1/18: Skyline AI recently took in $18 million from a Series A funding round led by Sequoia Capital which the startup will use to further build out their data science team and pay for that posh sales office they opened in New Yawwk.
Location, Location, Location
London-based Proportunity, which raised $1.7 million last year, might be an even more attractive acquisition for a company like Zillow in the future. Founded in 2016 by a pair of Romanian entrepreneurs, Proportunity claims its machine learning algorithms can accurately forecast which homes, and even neighborhoods, will experience the biggest bump in value over a certain time period. Proportunity CEO (and man with the coolest Star Wars-like character name) Vadim Toader, told Biznow (a trade publication for commercial real estate) that the startup’s AI platform works by analyzing historical pricing data against about 50 metrics on location and things like transportation, crime, schools, etc. It’s a process that’s almost as magical as Romania’s legendary Hoia-Baciu Forest. The algorithms then figure out what factors truly affect price, eventually moving to the stage where it can predict how prices will move. The machine is right about 85 percent of the time when tested against real past performance. Proportunity says it can get to 95 percent accuracy by the end of this year.
It’s All About the Amenities
Chicago-based Enodo took in $2.5 million at the end of last year for its predictive platform that serves the commercial multi-family (ie, apartment dwellings) industry. The algorithms do it all: They can identify attractive investment properties, especially those with some fixer-upper potential. Enodo’s machine learning models can also calculate market rents, identify the rental impact on any given amenity package, and offer statistical comparisons to comparable properties. In other words, Enodo can tell you if it’s more valuable in a given region to offer an in-unit washer/dryer combo or a community room with an S&M dungeon.
The AI of the Deal
Speaking of creepy: real estate agents. On the evolutionary food web, they fall somewhere between slug and Human Resources recruiter in our book. So we were glad to come across REX Real Estate, which has (mostly) replaced real estate agents with AI. Founded in 2014 just outside of Los Angeles in Woodland Hills, REX Real Estate has raised a total of $25.5 million, including a $15 million Series B in January. Investors included a who’s who of random big-company founders such as Sun Microsystems co-founder Scott McNealy, Best Buy founder Dick Schulze and Crate & Barrel founder Gordon Segal, among others.
How does a computer sell a house? Advertising. According to an article on CNBC, the machine shoots out an initial batch of online ads to buyers it thinks might be in the market for a new home. The algorithms then really go to work: as the first clicks come in, the machine figures out commonalities between those who clicked on the ad. It then takes those criteria and searches out new leads based on the patterns it finds. The platform also analyzes online behavior to identify potential home buyers and sellers. For example, someone who are making lots of trips to Home Depot might be fixing up a house to sell—or making pipe bombs. The company only charges a 2 percent commission fee, and has reportedly closed more than 200 transactions since 2016.
Not Yet Finished
A Finnish company called Blok is also automating the housing real estate business. The Helsinki-based company, founded in 2017, raised about $1.9 million in February. Blok is pretty niche, helping sell apartments in Finland through its online platform. It handles everything, from the paperwork to the photography. The AI part kicks in early during the process when the company’s algorithms pull housing data to produce a price estimate. The valuation is based on a variety of factors, such as the age of the apartment and its amenities, like being in close proximity to a herring fishery. In return, Blok gets 0.75 percent of the sales price. It currently has nearly 200 listings, such as these:
The startup has sold more than 250 apartments and says it saved customers about $1.75 million in fees so far. That’s a lot of karjalanpiirakka.
The Price is Right
AirBnb is worth about $30 billion and today pretty much dominates the home rental real estate market. That means there’s at least a few startups willing to compete. One of them is Beyond Pricing, founded in 2013 out of San Francisco. The company has raised $3.5 million and even acquired one of its competitors, a Brooklyn startup called Smart Host that had taken in less than $400,000. The combined company uses algorithms to find the best price for vacation rentals on any platform, including AirBnB itself or competitors like VRBO. Beyond Pricing claims its users make up to 40 percent more in additional booking revenue.
Chatting Up Customers
This listing wouldn’t be complete without an AI chatbot or two. Founded in 2015, Austin-based OJO Labs has raised $26.5 million, including a $20.5 million Series B this month. OJO uses natural language understanding to provide customer service on behalf of a real estate agent or brokerage. OJO can answer questions or help with home searches. For example, a customer might ask if a certain property has a backyard. It not only can confirm that there’s a backyard, but add extra detail such as the fact that there are four large oak trees that provide plenty of shade in the summer. Its image recognition software can also find housing matches based on a text conversation about open floor plans or S&M dungeons.
Founded in 2015, Montreal-based Roof.ai has also developed a real estate chatbot. The startup has managed to raise $350,000 in a series of convertible notes. Again, the idea is that customers can ask the chatbot questions and get specific results returned on the kind of homes they are interested in buying.
Moving On Out
Founded in 2014, First.io out of Durham, North Carolina has amassed $7.4 million, including a $5 million Series A just this month. Its AI trick? First claims it can predict which contacts within a real estate agent’s network are ready to sell a home based on a variety of factors, such as spending patterns, employment changes and income history. Apparently it has these details on 214 million people:
The company claims its platform is used by more than 200 agents at major brokerages around the country, saying its algorithms improve the chances of a broker connecting to a potential seller by four-fold.
Conclusion on Artificial Intelligence in Real Estate
A couple of conclusions immediately come to mind. First, the use of artificial intelligence in real estate is still very much in the early years of development, with only one company in our list at the Series B stage. There’s also a wide range of use cases, meaning there’s plenty of room for more competition as the tech improves and the industry warms up to the idea that machines are far more likable than humans when it comes to dealing with human beings.
During the course of our research we found that Zillow actually relies quite heavily on AI to produce its estimates. ZDNet did a deep dive into the technology last year and found that the company has dramatically improved its algorithms since it was founded about a dozen years ago. It now has price tags on more than 110 million homes, with an accuracy rate of about 5 percent. Since going public in 2011, Zillow has done well for its shareholders, returning +380% for the company’s Class A stocks, according to Google Finance. So, if you’re looking for some pure-ish play on artificial intelligence in real estate, you should take a closer look at Zillow – or subscribe to our newsletter and we’ll do it for you in a coming article.
Are you paying too much in transaction fees to your broker? Check out a brokerage firm called Zacks Trade that's offering $1 trades until 2019. After that, you'll pay just $3 a trade or a penny a share, whichever is greater. It's one of the cheapest brokers out there and you can also trade stocks on foreign stock exchanges. Click here for $1 trades until tax day 2019.