Investors now have more than 131,000 houses in Las Vegas Valley.
Las Vegas Review journal | Tribune News Service | Getty pictures
A version of this article was first published in the CNBC real estate newsletter with Diana Olick. Real estate game includes new and developing opportunities for real estate investor, from individuals to risk capitalists, private equity funds, family offices, institutional investors and large public companies. Register to get future editions directly into your inbox.
John Carrafiell, Co-CEO of BGO, a global real estate investment manager with a managed assets of $ 89 billion, is very proud that he sits right next to his full-time scientist.
The investment strategy, regardless of the market, has always rely on research and data, but artificial intelligence has brought this to a completely new level that changes investment research models and steroids developed a few years ago.
Carrafial, who has been working in the real estate business for around 40 years, said that he was increasingly frustrated about the research and data methods of the sector, which he really has not changed over the years. Everyone seemed to examine the same information and find the same conclusions. The question he said he kept asking himself: “How do we really exceed?”
He found that the answer analyzes all the past transactions of his company that go back 20 years, only use a computer model and remove the human element from it. The model found that the outperformance or under -performance by the local market, which was selected for the investment, was completely determined.
That may sound banal – since the mantra of real estate has always been “place, place, place” – the results asked his team to concentrate almost exclusively on the local market finder data when choosing its future investments, and not so much on ownership prices and national economic trends.
Get the property directly into your inbox
CNBC's real estate game with Diana Olick covers new and developing opportunities for the real estate investor and delivers in your inbox weekly.
Subscribe here to gain access today.
Of course, there are research companies that analyze and evaluate local real estate markets, but according to Carrafial, BGO made its results as something random. Instead, it looked at its own past and built up a model that was exactly what his best and worst achievement. The model contains all types of local market data points, including demographic and delivery trends that are unique for each location. AI then gave this model increased data volume and speed.
“We have taken thousands of data input, many that are free from the government. We have to buy many from telecommunications providers and great data. We found the key,” said Carrafiell. “And we know that it is exactly because we are examining it.”
BGO used his data science to inform a decision to invest in an industrial development in Las Vegas with partner Northpoint Development. Other data models indicate that this was not a particularly good investment.
Carrafial said that the “best research out there” showed that the investment in terms of performance and returns would be mediocre.
“But our model screamed, it will explode. We wrote $ 5.88 per square meter. We received rents in the 9 dollar range,” he said. “This does not happen in commercial real estate. It's no luck.”
The model, he said, saw that the inland Empire of California became too expensive and then analyzed the logistics routes. It found that instead companies could be large in Las Vegas, where both rents, taxes and work were cheaper.
“So they had an additional two -hour drive, but they saved 60% at their total costs, and that was exactly what the model saw,” said Carrafial. “The tenants we have there serve an entire region. They do not serve in Las Vegas.”
BGO carried out similar analyzes for investments in Florida and Rust Belt, which led to great returns in his investments.
“We believe that our performance has increased significantly due to this model,” said Carrafial.
But he admitted that the accuracy of the model is dramatically improved by artificial intelligence, but can never be completely accurate and hypothetical: “Boeing can pull out of Seattle, and the model cannot predict, right?”
While the BGO investment team focuses on the excitement models for potential properties, its loan looks at the downward modeling, since its risk lies.
The new iterations of the research model in the course of the street include asset assignment in various sectors of commercial real estate. The model would ideally suggest an optimal portfolio mix. The possibilities are still growing, which is why Carrafial says that he was chosen in the data like never before.
“Ai is an enhancer and a accelerator pedal that enables us to do so much more, but it's really data science,” he said. “It is [like] A six people, dedicated data science team who sits next to their CEO and next to their asset management and acquisition team. “