Why Opendoor is Misunderstood

I want to explain why I think Opendoor is such an interesting company. 

The most interesting thing about it is that it’s mispriced. It’s mispriced because people fundamentally misunderstand the company and its business model. They’re judging it by the wrong metrics. 

Which is understandable. The main thing that Opendoor does is buy and sell homes. So people compare them to other companies that buy and sell homes - companies like BlackRock. 

All the other companies that buy and sell homes have two ways of monetizing - either they charge rent, or they sell the home at a mark-up, often after holding it for years. Many companies do both. 

Opendoor doesn’t charge rent, and they don’t buy homes as investment assets. That’s because Opendoor’s not an investment company and it’s not a REIT. It’s a marketplace. 

What marketplaces do is increase the volume of a particular type of transaction (usually by making the transaction cheaper and/or more convenient) and take a percentage of each transaction as a service fee. 

That’s how Uber, DoorDash, and Airbnb work also. The thing that makes those companies different is that their transaction risk is very low. Once Uber connects a driver and a rider at the right time and place, the deal is virtually guaranteed to go through. The buyer will pay right away, Uber takes its cut, and Uber pays the driver the agreed-upon fee. 

Real estate transactions are different. The average sale price of a home in the US is $430,000. That’s about 3,300x as large as the average Airbnb homestay ($130). Real estate transactions take months and traditionally involve a whole cast of independent economic actors - agents, inspectors, contractors, mortgage lenders, competing buyers/sellers. Simply connecting a buyer to a seller does not create substantial value in home purchases because the transaction risk is so high. 

Opendoor’s insight was that they could eliminate transaction risk in real estate transactions by simply buying and selling the homes themselves.

In a traditional selling process, the seller will have to talk to a few real estate agents, get an inspection, do some repairs, maybe renovate the bathroom, clean the house and keep it immaculate, hire a staging team, shove all their belongings into one shameful closet to hide it from the strangers walking through their house and commenting on the floorboards and the family photos…

A seller who’s selling to Opendoor has a much simpler process. They can do a brief video walkthrough, send the video to Opendoor, and get an offer. The seller doesn’t need to do repairs or renovations. They don’t even need to clean the house. And they can close in days. 

Sellers love Opendoor. It’s NPS tends to hover around 80 - up there with the best brands in the world.

Because Opendoor is a marketplace and not an investment company, their next step after purchasing a home is not to rent it out or hold onto it while it appreciates. The next step is to sell it as soon as possible for something close to the purchase price. 

They can get away with reselling at the purchase price - or even lower! - because Opendoor has vertically integrated most of the services that surround a home transaction. Typically, around 10% of a real estate transaction goes to various service providers. Opendoor is bundling more and more of those services, which allows the company to charge (currently) a 5-9% take rate from each transaction and provides accelerating advantages in product. 

Already, when you purchase a home from Opendoor, Opendoor can provide mortgages, title, escrow, and home upgrades. Before long, they’ll have moving services. Opendoor is hurtling toward a product experience where on a Tuesday night, you can scroll through the Opendoor app, buy a house you like in Tennessee, sell your house in Colorado, and be fully moved in, with finances sorted out and a home upgraded to your specifications, in 3 weeks. And Opendoor will have handled every part of that process. That’s where Opendoor is going, and there’s no one else even close. 

When you see Opendoor as a marketplace that’s focused on creating the best possible transaction experience, rather than a real estate investment company, the most frequent concerns you hear about the company start to seem misguided. 

The big one is that the company isn’t profitable. 

The rap on Opendoor is that they’re a perpetually low-margin company, and the fact that they couldn’t turn a profit in these last few years, when home prices have been rising fast, is strong evidence that in the inevitable real estate market downturn, the company will lose a lot of money and maybe even go bankrupt. 

In evaluating a company’s growth and profitability, it’s important to understand where the company is in its lifecycle and what market they’re in. Opendoor is a well-capitalized, 8-year old company that’s growing like crazy. They’ve averaged more than 100% revenue growth for the last 5 years, and they’re expecting 100% growth again this year, to close the year with $16B in revenue. They’re expanding to new markets and launching new products well ahead of schedule. Opendoor has already demonstrated that they have profitable unit economics, which benefit from scale. They’re operating in a fragmented multi-trillion dollar market where they’re already the fastest-growing and one of the biggest players.

It would be truly bizarre to advise a company that’s accelerating into one of the world’s biggest markets with good unit economics and $2B in cash to shoot for a profitable quarter. 

Most of the time, in the short term, profitability and growth are in conflict. 

Opendoor shouldn’t aim for profitability for the same reasons that Facebook, Uber, and Amazon didn’t when they were young companies. When you have a fast-growing product that delivers a differentiated customer experience and benefits from network effects, you don’t step on the brakes. For Opendoor, choosing growth now means more markets, more customers, happier customers, more products, and more investment in future efficiency. Choosing profitability means less of all of that.

Growth drives down future costs and increases future pricing power. Growth today creates profitability tomorrow. It doesn’t work in reverse. 

It’s hard to recognize Opendoor as a marketplace because the company’s financials look different than other marketplaces. They look different because Opendoor takes ownership of inventories, so every transaction gets recorded as Opendoor revenue rather than as GMV. 

Functionally, Opendoor and Airbnb both facilitate a transaction and charge a take rate of 5-15%. But nobody mistakes Airbnb for a company with 13% margins. That’s because only Airbnb’s take rate is recorded as revenue. The rest of the transaction is recorded as Gross Bookings. If you list your Airbnb for $70, Airbnb will charge a guest $80 and keep $10 as a transaction fee. The $10 is recorded as revenue, the other $70 is recorded as Gross Bookings. Airbnb’s margin is whatever the company gets to keep of that $10. 

Let’s say Opendoor buys a home for $400K then sells it for $400K, and takes a 7% take rate on the sale and nothing on the purchase. Opendoor will book $428K in revenue, and the Cost of Revenue will be recorded as $400K plus the costs of servicing the transaction. Let’s call that extra cost $12K. 

The way this would be recorded in Opendoor’s financials is $428K in revenue and $412K in Cost of Revenue. That’s a profit margin of about 3.7%. Tiny! 

Alternatively, if Opendoor’s accounting matched Airbnb and other marketplaces, this transaction would be recorded as $28K in revenue and $12K in cost of revenue. That’s a profit margin of 57%. 

If Opendoor’s thesis is correct, and Opendoor can indeed price homes accurately enough to reduce inventory risk to manageable levels, then the framing of Opendoor as tech-enabled marketplace with software margins is probably closer to correct than how the market currently views the company.

Can Opendoor really price and sell homes accurately? 

This is of course the question that Opendoor started with. The very first thing they tried to prove. 

First they proved that they could do it in just one city, Phoenix, with just one category of houses - homes between $100,000 and $250,000 that were built within a certain timeframe. And then they proved it over and over, in more cities, with more kinds of houses, in more price environments. Opendoor did $8B in volume last year across 45 markets. You’ve gotta wonder when people will stop asking if they can price and sell homes. 

But of course real estate markets do go down, and the US residential real estate market looks poised for a crash. The market seems to think there’s a good chance Opendoor will crash with it. I think that’s very unlikely. 

Real estate crashes are slower and more diverse across geographies than people tend to think. In the housing crash from 2007-09, it was rare for home values to decrease by even 2% in a month. Opendoor sells most of its houses within 4 months, and they’re diversified across geographies. Once a downturn is identified, Opendoor can probably unload a large portion of its inventory without significant losses. And because Opendoor is a vertically integrated service provider, they can offset losses from home price depreciation with transaction fees. In fact, they’ll probably take profits on plenty of homes that they sell at a loss. 

Opendoor has a lot of other tools at its disposal. It’s the most active, sophisticated trader in the market. The company buys and sells 10,000 homes a quarter, has $10+ billion in lending capacity, and has the best-resourced data science team in the market. It wouldn’t be a surprise if Opendoor saw the downturn early and handled the selloff more effectively than others.

An appreciating home market is a good market for sellers and a depreciating home market is a good market for buyers. Opendoor is the biggest seller and the biggest buyer. During a depreciating market, whatever downside there is for Opendoor, there’s just as much upside. After unloading its homes, Opendoor can tighten up its underwriting criteria, make 10x more offers than any other market participant, and buy houses on the cheap. 

When you listen to an Opendoor earnings call, half of the questions are things you would ask a real estate economist. What are rates gonna look like next year? How long do you think inventory will stay on the market next quarter? How are mortgage rates going to affect homebuying? It’s a symptom of not understanding what’s important about the company.

It’s kind of like asking Uber about the price of gas and vehicle efficiency standards and inflation. Sure, some of those things may be inputs in pricing models that matter to Uber, but if you’re evaluating the company based on short-term changes in gas prices and interest rates, you’re missing the things that matter.

Over time, I suspect it’ll become easier to see Opendoor as a marketplace and to compare it with other marketplaces - Airbnb, in particular. There are some ways Opendoor comes out better. The market for real estate transactions will certainly be larger than the market for homestays in the foreseeable future. One thing I’ll be paying attention to is whether Opendoor will become a consumer breakout like Airbnb has.