"Real estate is by far the most screwed up industry in America," Glenn Kelman declared in a 2007 interview on 60 Minutes. Kelman, the CEO of the online real estate company Redfin, was following the familiar script of a disruptive internet startup. "We feel like things that Amazon or eBay or Yahoo have done for other industries, we can do for the real estate industry," he said.
Redfin showed homes for sale on an interactive map — a fairly new concept in 2007. And it had an equally novel business model: a do-it-yourself real estate service that let customers buy houses for a third of the cost of a conventional agent.
You know how this story is supposed to turn out: The startup upends the old way of doing things, driving established competitors out of business. But that's not what happened. Over the next few years, Redfin was forced — slowly and reluctantly — to operate more like a conventional realty company. As Redfin beefed up its service, it charged more too. Today, Redfin owes its success as much to adapting to conventional real estate practices as to defying them.
When I met with Kelman at Vox headquarters in April, he spoke with the same quiet intensity as the man who denounced the real estate industry on national television eight years earlier. He still believes Redfin is changing the real estate industry for the better — and that Redfin's unique business model allows it to do things other technology and real estate companies can't. But the gospel Kelman is preaching today is more nuanced than the David-and-Goliath tale he told nearly a decade ago.
"I can't walk around with a swagger and talk about destroying an industry," he told me. "I am more confident than ever that this change that we're creating is going to continue and accelerate. But if that means that some other real estate agent at some other brokerage has a tougher time making a living, that's not a reason for me to throw a party."
A lot of early internet companies, including Yahoo, Google, and Facebook, were online-only operations. But as the internet becomes ever more deeply embedded in our daily lives, companies that straddle the line between the physical and online worlds will become ever more important. Redfin doesn't have as many users as online-only competitors Zillow and Trulia. But in some ways it's having a bigger impact on the real estate industry.
"To change the game, sometimes you have to learn the game," Kelman told me. "To affect the real world, you have to be in the real world."
A different shopping experience
I don't just write about Redfin, I'm also a customer; my wife and I bought a house with Redfin in June.
I became interested in Redfin after talking to my friend Adrienne about her home-buying process for an article about home-buying last year. "Redfin was perfect for a type-A person like me," she told me. "Their agents are very knowledgeable and responsive, but they are not going to hand-hold you through finding houses to view."
As a type-A person myself, that sounded pretty good to me. I'm good at finding and organizing information on the internet, so I didn't feel like I needed a lot of hand-holding.
Yet our Redfin experience wasn't actually that different from the experience we would have had with a conventional realtor. We were assigned an agent named Thomas, who took us on our first home tour. He answered questions about the home-buying process and encouraged us to keep looking until we found a property we wanted to buy. Later tours were performed by other agents, but Thomas was always available to answer questions and guide us through the process of making offers.
If we'd been shopping for a home back in 2007, when Redfin first entered the Washington, DC, market, we would have gotten a more extreme do-it-yourself experience. Back then, Redfin charged customers $250 to go on a home tour — if they had agents available to show the property at all.
Back then, Kelman thought customers would jump at the chance to do most of the work themselves and pocket thousands of dollars in extra savings. He was wrong.
The early Redfin business model attracted a small, hardcore fan base, but it gradually became clear that it wouldn't go further than that. The per-tour fee stressed people out. Also, first-time homebuyers have a lot of questions that Redfin wasn't really set up to answer. When Redfin surveyed its customers, the feedback was clear: They wanted to be able to tour more homes and ask more questions, even if the service cost more.
Redfin shifts to a full-service model
"We didn't want to show houses because we knew it would take a decade to hire all these people," Kelman told me. "And we knew the shape of the business's profits would be a lot different."
The most successful internet companies exploit the fact that computers are cheap and human labor is expensive. For example, because almost all of Facebook's interactions with users are automated, a few thousand employees can serve a billion users. The more labor-intensive Redfin's service became, the slower and more arduous it would be to grow it and eventually turn a profit.
This is probably why Zillow and Trulia, which were founded around the same time and also displayed real estate listings on a map — never copied Redfin's brokerage model. They just ran websites and charged traditional real estate agents for referrals. That low-overhead approach allowed Zillow and Trulia to grow rapidly — today Zillow is the market leader with almost 10 times as many users as Redfin.
But by 2008 Redfin was already in the brokerage business. And it had become clear that the original, bare-bones brokerage model would never appeal to a mass audience. So in November 2008, as the real estate market was imploding, Redfin took the plunge.
"Redfin started life as a cult," Kelman wrote on the Redfin blog. "But our goal has always been to become a religion."
Shifting to a full-service model required hiring a lot more real estate agents. And that, in turn, would cost money. Traditionally, a buyer's agent gets 3 percent of the sale price. Before, Redfin had kept 1 percent of the sale price and rebated the other 2 percent to customers. Now it was keeping 1.5 percent, a 50 percent increase.
In 2012, Redfin beefed up its service again, hiring 50 more agents so it could reduce each one's workload and give agents time to get to know customers personally. Once again, the improved service came with a higher price. Today, Redfin often keeps more than 2 percent of a home's sale price, rebating less than 1 percent to the customer.
Buying a house isn't like ordering a book or an airplane ticket
Buying a house isn't like other consumer purchases. I can order a book or an airline ticket in minutes. By contrast, I spent months poring over housing listings to find the perfect property. And because I'd never purchased a house before, I had a lot of questions that I wanted a human being to answer.
And after almost a decade in the business, Redfin still hasn't automated many key steps in the home-buying process. For example, the company doesn't offer an automated way to get estimates of home values — when we needed guidance on how much to offer for a property, we emailed our Redfin agent for a personalized estimate. That's probably because publicly available data on homes is unreliable — Zillow offers a "Zestimate" of home values, but these figures are not very reliable.
Similarly, Redfin has done little to change the process of making an offer on a home. It does use a service that lets customers sign PDF documents online. But the documents themselves get filled out by Redfin agents and sent to customers by email — there was no way for me to prepare an offer on my own.
Ultimately, real estate has proved less prone to disruption than other industries that have been transformed by digital technology. At first glance, it strikes some people as outrageous that someone could earn more than $10,000 to help someone buy a house. But when Redfin offered customers a radically cheaper alternative, most of them didn't want it. Buying a home is the biggest financial decision most homeowners will ever make, and so most customers are willing to pay a premium to make sure they don't make a mistake.
Why Redfin has a better chance to disrupt home selling
In its early years, Redfin was more popular among buyers than sellers. Kelman says there are a couple of good reasons for this. One is that Redfin's original innovation — putting real estate listings on a map — was targeted at buyers. Also, sellers are older, on average, than buyers, and therefore less likely to take a risk on an untested new online realty service.
But more recently, Redfin has attracted a significant number of sellers. And Kelman sees that as a big opportunity.
"When you sell a house, you're the one who gets to make the rules for how it's going to be sold," Kelman says. Sellers decide how homes are presented online, when and how offers are accepted, and more. The power of sellers meant that Redfin had limited opportunity to innovate when it was primarily a website for homebuyers. But now that Redfin is representing a lot of sellers, it gets more say in how homes are sold.
Redfin now discounts more aggressively on the seller's side of the market than on the buyer's side. The company generally takes a commission of more than 2 percent when it represents a buyer. For sellers, it takes 1.5 percent, and in Washington, DC, it is experimenting with charging just 1 percent — a third of the 3 percent fee many real estate agents charge sellers.
Redfin has also invested in technology to show 3D walk-throughs of homes it's trying to sell.
And while he refuses to provide details, Kelman hints that Redfin is working on other ways to improve the home-selling experience. He points out that in Australia, for example, homes are more often sold in auctions. It would be hard for a conventional realtor to conduct an online home auction, but Redfin's broad audience of potential buyers might give it a unique opportunity to experiment with different approaches.
"The business has become very explicable to investors"
Redfin may or may not be changing the real estate industry, but it's thriving as a business. In December, the company raised $70 million to fund expansion into new markets. That's on top of $95 million the company received in earlier fundraising rounds.
Redfin has long aspired to be a national brand, but until recently it was limited to a handful of major metropolitan areas. Redfin's labor-intensive business model meant that expanding into new markets was an expensive proposition, and it took a while to demonstrate that the investment was worth it. But now Kelman says they've done it.
"The business has become very explicable to investors," Kelman says. "Anyone can understand how long it's going to take us to build share, what the profit from that market will be. And once you have that understanding, it's very straightforward to get capital."
The latest capital infusion has allowed Redfin to add more than two dozen new markets since the start of the year. The company now serves more than 70 markets overall. Recent additions have included El Paso, Texas; Reno, Nevada; and Dayton, Ohio.
There's a danger that bulking up will harm Redfin's ability to innovate. The larger Redfin's staff is, the harder it will be for new innovations to percolate throughout the company's nationwide workforce, and the more challenging it will be to keep standards of customer service high.
But Kelman argues that Redfin's growing staff of real estate agents will let it innovate in ways that a purely virtual company couldn't. He draws a parallel to Uber and Lyft, two other technology companies whose services depend on large workforces. "You could argue that Lyft or Uber isn't disruptive because you still have a human being driving a car," he says. "Uber could have waited for self-driving cars. But they didn't, and I don't know many people who aren't glad."
We're used to thinking of the internet and the physical world as different domains. But in a world where everyone has the internet in their pocket, the biggest opportunities may be for businesses that have a foot in each world, using digital technology to make a conventional service faster, cheaper, or better.
Correction: I misattributed a real estate agent's critique of Zillow that appeared in the Washington Post to the Washington Post itself.