Rocket Companies CEO：解决住房危机的方法

Housing, in some sense, for me is like that final frontier of fintech. It's going to be the end goal for most consumers. All of fintech, in some ways, leads to a consumer caring fundamentally about generational wealth. And generational wealth comes from things like homeownership, right? It comes from long-term appreciation. It comes from creating something that's safe and sustainable for you, your family, and your family's family. And that's the American dream.

Part of the problem is that all the old people have all the money. It is a catastrophic issue right now. If you're getting paid in cash, you might get a 3% salary bump every year, but the S&P 500 compounds at 10% a year.

Alex, in 2010, the median age for a homebuyer was 30 years old. Now it's 38. Why did that happen, and what can we do to change that?

Well, I think part of the problem is that all the old people have all the money. This really is a catastrophic issue right now. If you look back, the American dream has been homeownership for a very long time. The example that I like to point to is after World War II. Have you heard of Levittown?

No.

So you've never heard of Levittown? Surely you know of Levittown. This was one of the first tract-housing communities. Think about what Henry Ford did for the automobile. How could you buy a car before Ford? These were handmade things that were very, very expensive. I think it was James Levitt, who actually turned out to be somewhat of a racist, but that's a different topic. Levittown was the first, “I'm going to bring the Henry Ford factory to housing.”

It was actually a town in New York, built with, I think, thousands of homes, and sold to GIs who were coming back. Homeownership was just—you had lots of land. You had a population that was half the size that it is today, so it was just supply and demand. You didn't have all the old people holding all the money. You had this giant population that wanted homeownership, and you had lots of supply that was being built. There were also new, innovative techniques for building.

That's why I like to point to Levittown. What has happened since is, if you just look at the pyramid of who owns what, you've got to think about inflation as being 2 different things. You've got the CPI, where the Bureau of Labor Statistics comes up with a basket of goods and services. Gas is one, bread is one, and fish, eggs, butter, and things like that are others. They get more expensive or cheaper.

But then there's this other thing that I would call asset-price inflation, which is not actually part of the CPI. Part of the CPI is rent or perhaps the mortgage payment that effectively allows you to live in your house. Asset-price inflation is very, very different.

We wrote a piece a little while ago showing that the price of housing in the Bay Area has declined massively in the past 25 years. You'd be like, “No, no, it hasn't. It's gone up.” If you price it in assets, right? If you already have one asset—one asset would be stocks, or one asset would be that your parents left you a lot of money and an old house—then, because of supply and demand, there's only a fixed amount of real estate. There's only a fixed amount of Apple shares. Those have gone up in value relative to the U.S. dollar. These are the people who can buy homes today, and they are predominantly and disproportionately older.

It is really hard for younger people to afford their first house because they get paid in cash. If you're getting paid in cash, you might get a 3% salary bump every year, but the S&P 500 compounds at 10% a year. If you're buying with a basket of assets—and obviously you're not paying for your house with Apple shares—but if you happen to work at Apple, houses have gotten a lot cheaper in the Bay Area in 25 years. If you happen to work at a company where you have no ownership in anything and you're just getting paid a salary, house prices have gotten a lot more expensive.

There are many reasons why I could answer your question, but I think fundamentally that's the main one. Number 1 is supply and demand: We don't build enough homes. That's very, very different from the Levittown era, when it was, “Wow, we're going to build a shitload of homes.” That really happened in the 1950s.

It's supply and demand, and it's also this kind of asset-price inflation, which has created this tale of 2 cities for people who have assets and people who do not.

Is asset-price inflation the best explanation for how old people got all the money, so to speak?

Yeah, basically. It was much easier to build 100 years ago than it is today. This isn't about housing, but let me pose this: How long did it take to build the Empire State Building? What do you think? A year? What do you think, Varun? How long did it take to build the Empire State Building?

5 years.

110 days from start to finish. I assume today it could never happen. If you wanted to change a windowpane, it would probably take 2 years right now.

The other thing is that it's not just that the old people have all the money. That's a problem, but it was much easier to build things when the old people didn't have all the money but bought a lot of these properties.

Now it's just much, much, much harder to build. That's number 1. If you could just go build 10 million homes tomorrow, what do you think would happen to the price of homes? Would they go up or down?

Pretty sure they would go down.

Unless every economist is wrong about supply and demand, with the intersection being the equilibrium price. The main bottleneck there is regulatory?

I mean, it's regulatory, but it's not always regulatory. It's this term NIMBYism: “not in my backyard.”

Imagine that I bought a house, and I know that the old people who own the house next to mine have made so much money by owning it because they bought it in the 1960s. This is actually a true story. My first house was in Palo Alto. I bought it for, I think, $2.1 million in 2008—very perfectly timed, before Lehman Brothers fell. I figured I wanted to buy a house before the world fell apart. Not so much. I should have bought it 3 months later because it went down 50% in value.

My next-door neighbor was a retired Stanford professor. He bought the house, I think, in the 1960s. It had a bigger lot than mine, was worth more than mine, and he bought it for about $30,000. If he wants that $30,000 to turn into more, that's a natural thing. Everybody wants it to increase in value. If I buy something and it's not consumable, I don't care about this water increasing in value. I care about my house increasing in value.

If you build 10 million houses right next to mine, the house is going to go down in value. That's NIMBYism. NIMBYism then becomes regulatory, and you have people who vote for people who will give them what they want, which is NIMBYism.

It's not just that you have evil politicians who, out of nowhere, say, “We now need to take 10 years to go build something,” or that you can't build things. It's people as well.

Varun, what are your reflections or reactions to hearing this? What are your thoughts on how we can increase homeownership?

Yeah, I think Alex is right on the money. The reality is there's just a lot of things that have shifted over the past 20 or 30 years around housing. It's cultural shifts, higher home prices, and higher rates.

I read a cool statistic the other day: In the 1950s, the average size of a starter home was 985 square feet. If you look at the size of a starter home today, any guesses on what that is?

What is it? I have no idea.

It's almost 2,500 square feet.

Our cultural expectations around what a home is have fundamentally changed. But you also have these other generational dynamics. People today are settling down a little bit later in life. Alex talked about affordability, but some people want to buy and just can't clear that affordability hurdle.

There are other things that I think will improve as we think about the future. The cost of building a home has got to come down, right? Things like robotics, 3D printing, and advancements in materials science will help with that, but that's a little bit further out into the future.

The nice thing about this AI revolution is that it's not just that it will help us with all the traditional use cases you see today around ChatGPT, what Anthropic is doing, and agentic AI. It's more about applied AI in the context of robotics, advancements in 3D printing, and materials science.

Our culture has fundamentally shifted a little bit. The expectations around a starter home have shifted. But then you have to attack the problem from the other way, and that's where you need a little bit more of a paradigm shift in the technology space as well.

If you were to predict over the next 5 to 7 years how the process of homeownership changes, or how the statistics we shared around homeownership change, what would you expect?

Yeah, I think there are a couple of things that would change fundamentally.

I think you'll start to see the applications of AI be a little bit more geometric in nature. It won't be that a computer is smart at knowledge-worker-type jobs, but that it can handle manufacturing, building, and more process- and workflow-oriented physical tasks, right? Alex has a great analogy around bits problems versus atoms problems that I think Andreessen is very famous for pioneering. So I think you'll start to see that, and I think that will significantly change the manufacturing environment in general.

I think the other thing that you'll also see is that part of the problem with home ownership today is that the reason it takes so long is there's just a lot of work, right? Providing your data, providing your documents—it's a gigantic qualification process that involves complicated terms like underwriting and qualification, money movement, et cetera. Why is it so easy to walk into a grocery store and buy a chocolate bar with a credit card, and yet the process of buying a home is just a slightly more complicated version of the same thing?

I imagine that in 3 to 5 years, a lot of that workflow gets hyper-compressed. A consumer, their financial readiness, and their qualification criteria are just things that happen in real time. I think that's exciting because, first, the amount of effort required to know whether you qualify for something as expensive as a home purchase can be significantly compressed. Second, you start to see more leaps in technology, where the costs of building, manufacturing, and servicing a home come down.

I think if we get that right—and this is something that I think the administration is rightfully focused on—we can start to see more new homes built at a faster rate, right? To Alex's point earlier, how do you get more inventory on the market, which would create more price pressure so that even if mortgage rates are elevated or high, at least the cost of the home can come down? I see these intersection points starting to influence the process, hopefully for the benefit of the consumer. Alex, anything you'd add to that?

Yeah, I mean, the technology thing—I was quickly ChatGPT-ing to refresh my memory on Levittown. This was the first planned suburb, in 1947. These were all returning GIs, and it was like, “We need to build housing for all of these people that came back.” What did they do? Remember the famous picture after World War II where the GIs like kind of swooping this woman down and kissing her? What did they do after that? They got married, they bought a house, they had 5 kids, and hence the baby boom.

They needed to live somewhere, and apparently it was a 985-square-foot house. A lot of these were constructed so quickly, and the technology wasn't bad at all. It was like, “Let's bring the Henry Ford factory to housing.” That was Levitt & Sons, William Levitt. They did some bad stuff, too: They famously had a rule that you could not rent or sell to somebody who was not white. That part was obviously bad, but the good part was, “We are going to build a shitload of homes, and we're going to do it in a smarter way.”

If you have an assembly line for homes—and Lennar and others do a version of this—when they decide to buy a giant tract of land and develop it, they don't just build 1 house, finish it, and then build the next house. It's really cool watching these communities get built because it's like, “Okay, today is foundation day,” and it's foundation, foundation, foundation. They just do all of these things. Then it's, “Okay, today is framing day,” and—boom, boom, boom, boom—these things go up very, very quickly.

A lot of homebuilding is, as you said, Varun, the atoms-versus-bits thing. It's just really hard to construct things, but we do have solutions to this. Modular housing is not a pipe dream. You can do this. There's this very cool, famous Chinese woman—it's almost like a meme—who's showing off an RV home: “Look, shower. Look at this.” It's pretty funny. It's this very popular TikTok and YouTube Short thing.

I think you have 3 main things. You need to make it as easy as possible for people to buy a house. I've never done this before. I buy a house once in my lifetime. What do I do? All I know is that I pay my rent every month. I know how that works. I'm scared by this process of home ownership. Where do I buy? Do I get outbid? Do I get a mortgage? This sounds so complicated. I'm going to keep renting.

So many people who rent do want to buy. This is not surprising. Would you rather—the American dream is to own a house. You currently rent. Do you want to be part of the American dream? Everybody's going to say yes. How do you make it as easy as possible? It's making the process of buying as easy as possible, and it's making it as affordable as possible.

When you get to affordability, it's probably the financial side, which is the mortgage. The United States is actually quite unique in that most countries don't have 30-year-term mortgages to repay your principal and interest. Sometimes you'll have a shorter duration. But you have the financial side, and then you have the construction side. I know more about the financial side than the construction side, but if you make both of these cheaper, guess what? More people will want to buy. You're going to open up this aperture.

The other thing that I would add is that I gave a presentation about this a long time ago. I never thought it would be popular to give a presentation on home ownership and housing and all these things that are happening, and somehow that was a popular little video that I made. I think there needs to be less of a binary between either I rent or I own. That was it.

A lot of these innovations that have popped up—what is Airbnb? You have this piece of real estate, and what do you do with it? Again, you can rent it or you can own it. You only have 2 options. No. You can rent it for a month, or you can buy it and save money—or rather, help make your payments—or have a second source of income or whatever. The Olympics are coming to LA in 2028. If you have a little apartment near the stadium, go rent it out for a month. That's an innovation that has actually made home ownership more affordable.

We were an investor in a company that was helping do rent-to-own. You rent, but you're basically setting your rental payment on fire every single month because it doesn't help you in any way, shape, or form in terms of getting ownership in a house. What if you could get a house that you have the right to buy later? That would actually bring down housing costs to a certain extent, because there's a saying by Warren Buffett that I love: Nobody pays to wash a rental car, right? That makes sense. Why would you wash a car for Hertz? You're returning it.

If you know that you're going to buy your rental, you'll probably take better care of it. If you take better care of it, the landlord doesn't have to worry about as many things, and then the cost goes down. There's a lot of in-betweens that I think technology—and, more importantly, entrepreneurship—can help drive. But you need a regulatory environment that allows it.

Are you excited about the fractionalization experiments or these other forms of financial engineering? Do they solve real problems? Are they feasible?

Can you define that?

There are some startups that have been trying to say, “You can own part of a home.” We can just increase the number of people who own part of a home.

Yeah, I mean, there's the blockchain stuff that doesn't really make sense to me. I'm a big fan of crypto, but crypto exists for things that are purely digital. If you're trading things that are purely digital, it works great. If you're saying, “I'm going to represent something that's physical, where I need to assert my ownership”—if somebody's living in my house illegally, that's called breaking into my home. I call the police and say, “I live here. Eric broke in. Kick him out.” You can do that.

But if I say, “No, no, no, on the blockchain it says that Eric—” No. The guys with guns enforce the laws, and they look at who owns the property based on the county recorder's information. Should it work that way? I don't know. That's how it works.

That stuff I'm not sure about. But, yeah, we're an investor in a company called Point that allows you to sell part of your house. You have a lot of people who are house-rich and cash-poor. Again, that's another thing that doesn't make sense. If I have $50,000 in credit card debt and a 620 FICO score, but I bought a house in 1950 and I own 100% of it, that doesn't make sense.

I shouldn't have to sell my house so that I can pay off my credit card. Why don't I sell 10% of my house? Stuff like that just takes a binary of either I rent or I own, or I either own all of my house or none of my house.

Like, you can get a little bit more creative around the edges by giving people more options. I want to segue and go deeper into mortgages. Alex, you’ve been investing in fintech for a long time and built a company in this space. How is the mortgage tech space viewed? How do mortgage and fintech work together?

I think what’s really interesting about mortgage is—I’ll tell a story by way of background. Here’s my little parable or story. I get to college, and I don’t have a credit card. I went to Harvard, and there’s a store called the Harvard Coop—or it’s a co-op, but it’s called the Coop—and that’s where you buy your textbooks and everything else. So I want a credit card, and they’re like, “Apply for a Harvard Coop credit card and you get a T-shirt.” I was like, “Woo, I want a T-shirt. That sounds great.”

So I applied for this card. I had, I think, a $75 credit limit—a lot of credit extended to me. It was a bank called First USA, which eventually became something else, which eventually became something else, which I think eventually became Chase. I got my card, and they gave me a free T-shirt. Why did they really want me as a customer? They didn’t think I was a supermodel and want to put their T-shirt on me. Why did they give me the T-shirt and the card?

They were betting on—we talk about this a lot—CAC and LTV. The customer acquisition cost was a T-shirt. The lifetime value is not how much money I’m going to make for them at 18, when they only extend me $75 of credit. The lifetime value is that, if I stick with that bank, eventually I’m going to make a very, very valuable transaction.

Banks make money on net interest margin. They take deposits, they make loans. What kind of loan might I, as an individual, take out with this bank that would generate a lot of value for them? Probably a mortgage. So the key inflection point for a bank, or a lot of financial services companies, is: I now have Eric or Alex or Varun as a borrower. I’m going to make a lot of money on them, but I kind of have to—I can’t just show up at the 11th hour and say, “Hey, I know I’ve never met you before. I’ve never given you a free T-shirt. You have no financial relationship with me whatsoever. Here’s a mortgage.”

You’re like, “Nah, I’m going to get it from the guy that gave me a T-shirt before.” Or I’m going to get it from my real estate broker. But the lifetime value—it’s funny, in finance, LTV normally means loan-to-value, right? The value of the house is $2 million. I got a $1 million loan. That’s 50% loan-to-value.

But in most of startup land, when we talk about LTV, we talk about lifetime value. The majority of lifetime value for a consumer, even an 18-year-old, is going to happen probably 15 years later, when they buy a house. That’s such a valuable inflection point, or when they get a HELOC. It’s a financial product where you are taking a large loan responsibly, you’re going to pay it back, but the amount of money made at that point in time is so much higher than when you’re an 18-year-old kid getting a free T-shirt with your $75-limit credit card.

Varun, what would you add to this intersection of mortgage and fintech?

There’s a great movie that came out, I think, around 15 or 20 years ago. It’s called Thank You for Smoking. I don’t know if you’ve ever seen that one.

Yeah.

There’s this one line that popped into my head where she kind of scams him, gets some information out of him, and he’s like, “Why did you do this?” She thinks for a second and says, “For the mortgage.” That line has always stayed with me because, in some ways, I’ve been around fintech for a long time. I was at PayPal, Groupon, and Intuit.

When people think about fintech, they think about personal loans, payments, investing, taxes, money movement—a lot of these kinds of businesses. But one thing I’ve learned just from talking to a lot of consumers is that all of those are a means to an end. Housing, in some sense, for me, is that final frontier of fintech because it’s sort of the end goal for most consumers.

Alex said it really well: when you think about renting, renting is part of a funnel that is a continuum toward buying, right? So all of it, to me, leads to housing. Housing is so important to the economy. It’s 20% of GDP, a $5 trillion market, and it’s very complicated. It’s fragmented, and it’s got a lot of moving parts.

The mortgage process itself is also really interesting. You have the concept of a loan: you have an originator that underwrites the loan and brings you to the closing table. Once it’s closed, you have the mortgage note itself, and that gets sold off to the secondary market, to these GSEs—government-sponsored exchanges like Fannie and Freddie. They use that to free up capital and maintain liquidity.

Then you have mortgage servicing rights, which is where you have an ongoing relationship with the mortgage as a consumer because it’s typically the biggest transaction any consumer is going to make in their life. That’s where you manage your payments, your property taxes, and your escrow.

What I think is really interesting is that all of these parts and processes have evolved to be very disparate. You have the home search and real estate experience, where you use websites like Redfin or Zillow to go through that part of the journey. Then you have the mortgage process, which is where you apply for financing and credit. You go through title and appraisal, and then you close. Then you go into servicing.

These are all completely different parts of the equation, which is why that LTV-to-CAC thing is so important. Typically, when you value a business or a company in fintech, or really any consumer business, you think about LTV-to-CAC. You think about, well, what’s the lifetime value of this relationship relative to the customer acquisition cost?

The problem with housing in general is that these, if you think about it as a funnel, are all disparate. A consumer essentially flies out of one funnel and into another funnel. So the economics are not great, given how big of an industry it is and just how big of a transaction it is.

That’s something that really fascinated me: all of fintech, in some ways, leads to a consumer caring fundamentally about generational wealth. Generational wealth comes from things like homeownership. It comes from long-term appreciation. It comes from creating something that’s safe and sustainable for you, your family, and your family’s family.

That’s the American dream. That’s what we thought about. But the economics of the business are incredibly disparate. So that’s why, for us, a lot of it has been around integration. It’s about more vertical integration. It’s about connecting these parts of the experience because we think we can, one, build a better experience; two, create a lot more efficiency and therefore lower the cost; but then three, just completely change the economics of the business as well and create something that is a bit more of a new species.

Mortgage is a really interesting business. I’m relatively new to it. I’ve only been 2 years or so in this space and in the industry, but it’s fascinating. It is so big and so fundamental to the economy. It is so important to consumers from a long-term perspective. And yet, there’s so much opportunity to modernize it, transform it, and improve on it. These are all the things that are exciting to me.

Let’s get deeper into Rocket and go into both its history and its future. On the history side, if I understand correctly, the broker model used to dominate, and then there was a transition to direct lending, and Rocket is one of the big beneficiaries there. Why don’t you give some of the history of how the model evolved and the story of Rocket?

Rocket, first off, has been around for a long time. This company is 40 years in the making, and that’s a generational thing. We have been around for a long time, and we’ve grown. We are one of the largest employers in Detroit.

We started by really transforming the mortgage experience from the beginning. I think we were the first to put mortgages on the internet. We were the first to put them on a mobile phone. We are now the first to really embrace the AI-driven version of the mortgage experience.

But a lot of this is about how hard it is to build this kind of experience at scale. The mortgage process is complicated. You have to build pricing, licensing, and hedging infrastructure that works on a day-to-day basis. Every state and every county has different lending requirements, different regulatory requirements, and a whole suite of different types of products: FHA, VA, 30-year fixed, and adjustable-rate mortgages.

The compliance requirements change—sometimes day over day, week over week. So, it's a gigantic workflow engine that we've built over the past 40 years. We're licensed in all 50 states and 3,000 parishes. We have this mindset of continuous innovation and continuous evolution.

We're great at what we do. It's why we have the most trusted brand and have become the largest mortgage company and lender in the industry. But I'd also say that we're very restless. One of our values is “obsessed with finding a better way.” That's something that drives thousands of our team members to do more restless, more disruptive things and continue to transform the space day over day, week over week, month over month.

I've had the fortunate benefit of being a new CEO. I've been in the role for 2 years, and I'm the first outside CEO in the company's history. But it's a very special company. There's so much potential for us to do more. I would say that we're quickly evolving from being a mortgage company to a homeownership company.

That means we're not just creating innovation in the mortgage space; we're really transforming homeownership, because that's what it's really about. It's not just about the financing aspect. It's about the search and real estate aspect. It's about servicing. It's about making a 30-year bet on consumers who are making 30-year bets on us.

Our grand vision is to evolve into a homeownership company. That's why some of the decisions we've made around acquiring Redfin and Mr. Cooper are fundamental to that thesis. We have a long legacy of building great products and experiences for our consumers. That's how we built a brand, and that's how we built our presence. We have thousands of team members who are passionate about what they do, and we have a very loyal team member and talent base working for us.

We actually had a celebration. We turned 40 this past year, and we had a very special experience for our long-term, tenured team members. We had over 500 team members who had been with the company for over 20 years. Wow. That's pretty cool.

I've been in Silicon Valley for some time. I've been in Southern California and Seattle, and you don't have a lot of companies that have that kind of loyalty. That's something that we really appreciate, because you guys know this, right? Innovation happens quickly, but it takes a lot of dedication and passion to actually create something special. You have to put a lot of time into it.

Sometimes, the first 2 years are experimentation, realizing what works and what doesn't work. In your third year, you start to figure things out. In your fourth year, you figure out better product-market fit. In your fifth year, you figure out distribution. What I love about Rocket is that we have a lot of folks who have been here a long time and will run through walls for this company. That kind of loyalty and dedication is something that makes us what we are, and it's something that we're betting on to transform us into the future as well.

Alex just hit 10 years at Andreessen Horowitz. I was also similarly inspired at the company picnic, just seeing a bunch of people who've been here for—

I got a gold watch. Yeah, if you're here for 10 years, you get a gold watch. So, I got a gold watch. It has a nice little engraving on the back.

To take a pithy take on this, a lot of companies in Silicon Valley are like, “I use this product every day, and the company can't figure out how to make money.” Then you have this other type of company where it's like, “I make so much money. How do I get the person to use me every day?” It's not because I just want somebody to like me; it's because I want to add value every day.

This is what I found so compelling. This is not part of my day job at Andreessen Horowitz. I invest in a lot of proptech companies and a lot of fintech companies. I have an extracurricular activity, which is Rocket. Part of the reason why I find it so interesting is that Rocket made—I think it was—$10 billion in net income in 2021.

Rates went down dramatically. People were like, “I'm paying 6%, or 5%, and I can refinance to 2.5%. I'm going to do that. I'm going to save a lot of money.” Rocket was at the epicenter of that. But you don't refinance your mortgage every day. You don't go buy a new house every day. These are not daily-active-use products.

Meanwhile, in Silicon Valley, you have all of these daily-active-use products. People ask, “How do we make money? We can't figure out how to make money. We get people to use this—like ChatGPT. People use it every day. Are they making money?” No. They're losing a lot of money.

Which of those 2 is more unique? Is it that I have a product that people use every single day? It's like a toothbrush. Remember, Larry Page had a rule at Google: “We will not launch new products unless they pass the toothbrush test.” What's the toothbrush test? You have to use it every day. Hopefully, you brush your teeth every day.

On the other hand, you have companies that don't pass the toothbrush test. But that's actually not a qualification for success, right? $10 billion in net income—that's success. I would rather start off with a very valuable thing that's a real business model, perfectly run and very profitable. How do you add more products and services that add more value to the consumer?

You go to the dentist twice a year. That's it. Is there a way to interact on a more regular basis? That's where things like mortgage servicing come in. You get a bill every month. That's an engagement. What do you do with that?

Most people—it's actually really interesting—there's a story that Tony Hsieh used to tell about Zappos. Zappos was one of the very few companies that decided to take customer support and try to turn it from a cost center into a love center. They wanted to really engage the consumer.

There was a famous story from Zappos where there was a woman, and the customer-support representative at Zappos sent her flowers because something bad had happened to her. You don't expect that to happen. Normally, it's like, “I hired McKinsey, and they said I could cut two-thirds of my people this way and that way. Then I could put people in this IVR, and I'll save $4 million a year.” That's how a lot of companies treat sending a bill to a customer or doing customer support with a customer.

Instead, look at it as an opportunity to upsell something—not in an evil way, but as an opportunity to say, “We have a monthly communication with you. What is it that we should do?” The answer can't just be, “We're going to drive down the cost to zero, have robots do everything, and you're going to hate us.” It's a really interesting opportunity if you're engaging somebody every month to do something with them.

You start off with the profit center. That problem has been solved. That's a really hard problem that entrepreneurs who come into our office every day are trying to solve. They're saying, “I'm going to figure that out eventually. But right now, I have a product that people use. I've got a toothbrush, but I have no idea how to make money from it.”

The other way of looking at this—and I think it's really interesting—is that it's a 40-year-old company. Varun's been there for 2 years. You have a real profit engine. How do we turn it into a toothbrush? You made a wave of acquisitions—high-profile public companies. What's the strategy for how this all fits together? What are you trying to do?

At the end of the day, we want to fundamentally redefine the category itself. Our thesis is very simple: if we connect more parts of the ecosystem, we can build a better experience. We can pass on that value to the client in the form of lower fees, lower costs, and less friction. We can build fundamentally different economics around the business and drive growth.

From our perspective, when you think about funnels, mortgage is, in some sense, one of the world's most complicated user-funnel products. When I was at Intuit TurboTax, TurboTax was another example of a funnel-centric product. Our strategy is pretty simple: we want to serve clients across the entire journey of homeownership.

Not just the financing and mortgage aspect, but also the top of the funnel—the home-search and real-estate aspect—and then, at the bottom end, the servicing aspect, where they have a lifetime relationship. We've made 2 acquisitions in service of that.

The first one is Redfin. Redfin is the most visited real estate brokerage site in the U.S. They have relationships with 50 million monthly active users. These are consumers who use the Redfin app and the Redfin website.

They search for homes every single day. What I love about Redfin is that they have an amazing mobile app. Most of those 50 million users use the product daily, and they use the product on a mobile phone, so they have that really nice, rich interaction.

They also have a network of thousands of real estate agents that we now employ with the company, as well as a partner-agent network. What that represents is really the start of the homeownership experience. Some people start with the mortgage and the financing. Other people start with the house—browsing, scheduling tours, exploring the home.

That’s one part of the experience that we want to connect. Those 50 million relationships really help us build a top-of-funnel relationship. The next thing that I think is also really important is that we want to connect that to financing, and then connect that into the servicing experience as well.

The reason that’s important is because it allows us to build more relationships with clients so that we can reservice them with products and services. Once you’re in a relationship with Rocket and you have a great experience with home search, real estate, mortgage, and servicing, the beautiful thing about it is that you create loyalty.

When that client is ready for their next homeownership transaction, it doesn’t have to be a new purchase. It could also be a home equity loan. There are trillions of dollars of home equity now trapped in consumers’ homes, and there’s a lot of equity there that you can use to generate cash flow.

To Alex’s earlier point, what are they doing today? They’re racking up more credit card debt. Instead, it’s probably better to take more equity out of your home. What better way to do that than with the provider that you already have a relationship with?

For us, the big realization that really led to these acquisitions is that these relationships are part of a super-funnel. They’re not a singular experience. If we can do a good job integrating those parts of the experience, we can create a better relationship with clients, create more loyalty, and be their lender for life. That’s the fundamental thesis.

Now, with Mr. Cooper and Rocket, we have 10 million clients in our servicing book. Those are clients that we have a lifetime relationship with, and that’s 1 in 6 mortgages in the U.S. So it’s massive scale and massive distribution.

There’s another saying where first-time founders really think about product-market fit, and second-time founders really start to focus on distribution. We have created a pretty amazing engine around refinance and purchase, but the question is: How do we now get more distribution? That’s really the thinking behind these acquisitions: We can now get more distribution.

The last thing I would say is, if you believe that the world is shifting into this AI world, one of the fundamental things that you need to make that successful is access to more data. That data allows you to build better models. Those models give you a better experience, but also a better understanding of consumer behavior. That just allows you to build a better experience.

It’s a pretty simple thesis: connecting more parts of the experience that are naturally part of the consumer journey and making it more seamless. There are fewer points of data entry, less waiting and wondering, faster turn times, better rates, lower costs, and better fee structures. It’s about earning loyalty over the lifetime of a consumer relationship.

These acquisitions are direct accelerants of our core strategy, and we’re excited about them. It’s early days—2 big public-company deals. The overall company is growing by approximately 60% in terms of its overall size. We now have more national presence, with locations in different parts of the U.S.

That has an ancillary benefit of allowing us to attract more talent to the company as well, and to build our team members in a way that allows us to be more innovative and grow and scale. That’s the core thesis.

Is it accurate to say you’re trying to vertically integrate as much as possible?

I think there are 2 things. One of the things that I’ve learned when it comes to large-scale M&A is that you want to be very intentional about when you are integrating and when you are accelerating. I’ll give you 2 examples, because I think the way we look at Redfin and Mr. Cooper is a little bit different around integration versus acceleration.

The word “acceleration” is really important, because I think sometimes what companies get wrong is that when they acquire a company, they try to assimilate it too quickly, or they try to assimilate it not quickly enough. They’re not intentional about why they’re doing what they’re doing.

When you think about something like Redfin, it’s a very successful brand that has a very successful following with a very loyal group of consumers. One of the things we talked about was that the Redfin brand is super important. Consumers have a lot of affinity for that brand because they built a great product.

As we think about the integration of Redfin and Rocket, we wanted to strengthen Redfin. We did not want to assimilate Redfin too quickly. We wanted to preserve the brand, strengthen the brand, increase the focus on traffic and demand generation, and invest in the real estate strategy we have.

One, it helps us build that top of funnel. Two, we don’t want Redfin’s brand to disintegrate. We don’t want consumers to lose that affinity. If anything, we want to make it stronger. We have deliberately decided to make Redfin stronger by allowing it to operate a little bit more autonomously, and that’s very intentional.

When you think about Mr. Cooper, the biggest synergy is the integration of the origination mortgage business and the servicing business, because we want to bring those 2 things together so that we can recapture and create more relationships that go between origination and servicing.

Mr. Cooper looks a lot more like Rocket. They have an origination business and a servicing business. With that company, we are going to rebrand. We are going to call it all the Rocket platform. We are going to fuse the organizations more closely together.

That is a much more intentional approach around integration versus acceleration. We’ve studied what has made acquisitions successful and what has made them unsuccessful by looking at a lot of patterns and practices, and we want to be very intentional about that.

To answer your question more directly, integration is our number 1 focus. It’s the number 1 focus across the company. We have very specific goals and very specific owners. We’ve organized around it, invested in specific milestones around the synergies, and we want to get this right.

This is a big bet that we’re making across the company. These acquisitions are billions and billions of dollars in value, and they’re very important to our strategy. Mr. Cooper is almost half the size of Rocket. Redfin is about half the size of Mr. Cooper. When you look at it in totality, it’s quite a substantial addition to Rocket that we’re making.

It’s a very important priority for me, for our board, and for our leadership team. I’m also excited about the talent. When you think about the Rocket leadership team as well as the organization now, we have many of the best and brightest in the industry.

We have people who are very steeped in mortgage and servicing, and in the technologies that surround them. We also have leaders like Alex who are now on our board and really represent the future state of the art in fintech and in tech in general.

We have a new CMO and a new CTO who come from industry, not mortgage. We have a very balanced leadership team, and I think these acquisitions are going to strengthen that significantly as well.

We were talking offline about the importance of companies being balanced in their business model. Why don’t you explain what that means?

One thing that is interesting about companies, especially in regulated environments or environments that are sensitive to macroeconomic dynamics, is that their business and profitability fluctuate. In good times, like low-rate environments, they print money. In other times, they tend to struggle. They float a little bit with the wind.

One thing that I’m really excited about with our company, especially with these acquisitions, is that we are now incredibly counterbalanced. We can survive and thrive in any market, rate, or economic cycle.

The reason for that is that our origination and servicing businesses counterbalance each other. For example, when rates rise, the value of our servicing portfolio continues to increase, and we earn recurring revenue on an increased servicing book. But when rates fall, we can originate more mortgages and refinance, creating new opportunities for that same servicing book.

The cool thing about Rocket is that it’s one of the only companies—and I would argue the only company—that is super-counterbalanced in the housing industry. It has the ability to survive, thrive, and grow market share in any market or interest-rate environment.

And that's something that I think is tremendously exciting when you think about a company in a multitrillion-dollar industry that has single-digit market share, where there's a lot of consolidation and a lot of opportunity to really disrupt and transform experiences with artificial intelligence. Having these different assets put together in the same place makes us very unique. I was going to say, there's a concept in math called a Fourier transform.

Basically, if you have any function—if you have a line—every business's revenue and profit should look like this. If you look like this, everybody's happy. If you look like this, everybody's not happy. If you look like this, people are probably not happy. But actually, there's nothing wrong with that. A Fourier transform basically says you can decompose any function into a bunch of sine curves.

If I have a line that looks like this, it might actually be composed of 50 things that look like this. But to Varun's point, they counterbalance. A big mistake that entrepreneurs often make is, "I have a business that is a sine curve. Let's kill it. I don't like that business." No, no, no—it's actually great. Most businesses are, in some way, shape, or form, cyclical, even if you don't think of them that way.

Add another sine curve with a slightly different period so it counterbalances. The banks have actually done this. JPMorgan is the most valuable bank in the world, and they've done this because they have a bunch of sine curves. Should they get rid of their investment banking business even though, in 2022, the investment banking business was not profitable? No. They're going to make a lot of money sometimes. It's a feast-or-famine type thing.

But let's add wealth management, because we get 50 basis points on $1 trillion every single year. Let's add retail. They have so many different product lines, and the reason is that they're all—if you have a curve that looks like this and then the exact opposite curve, you combine this with that and get a straight line.

You want straight lines because they're predictable, but your business, under the covers, typically is not a straight line. It's composed—it's a Fourier transform. You have these overlapping sine curves in such a way that you hopefully do get a monotonically increasing straight line.

Okay, you were talking earlier about money-printing machines and Silicon Valley startups that have toothbrush tests but don't have a money-printing machine attached to them. Rocket has been able to acquire a sort of demand machine through Redfin and others. Why is it so hard in real estate, if you own the place where people are searching every day, to build the money-printing machine next to it?

Well, there's so much latency. Zillow is a good example of this. A lot of people use Zillow, but it actually doesn't make that much money because it's a lead-generation machine for agents. That's how they make money. The vast majority of their revenue comes from—you do a search. Why are you doing a search? You're not even doing a search with purchase intent.

Google makes a lot of money. We've talked about Google a lot on many podcasts, right? Google makes lots of money, and it's a freemium model: 99% of the time, you're searching for, "How do I kill this annoying fly that keeps buzzing around our pod?" And 1% of the time, it's, "I want to go buy a fly swatter." That's purchase intent. And there is a fly in our studio, Varun. That's what I'm referencing here.

So 1% of the time there's purchase intent. There are people who—you know the term "Netflix and chill," right? There are people who have done this with looking for homes. It's, "I'm going to spend all night looking at homes that I cannot afford in a weird part of the world that I'm never even going to visit." Why do people do that? I don't know. It's fun. People like looking at homes. It's aspirational.

"I heard that Sheryl Sandberg sold—let me look at that house." Or, "I heard that Sharon Stone did this," or, "I heard that Liam Neeson did that." You're not going to buy Liam Neeson's house, but you're still curious. You go look at it. The purchase intent tends to be low.

You have people—it's kind of like there are some search engines that are very tightly coupled to your transaction, and real estate is not always that way for 2 reasons. Number 1 is that there's entertainment value to it, and number 2, there's a lot of latency. I was looking at homes when I couldn't afford one, when I was just a little renter in Santa Clara, having just moved here. I couldn't afford a house in Palo Alto, and I would look at homes all the time. There was purchase intent, but with years of latency.

Compared with Google, that's not how you use Google. It's, "I want to buy a fly swatter." I type in "fly swatter," and then I see 50 search links pop up. I click on one and buy. Maybe I'll buy tomorrow, but there's much more near-term proximity to that.

That's part of the reason why you've got a DAU product, and that does not necessarily mean you can bolt on a monetization engine and get it to work. Particularly when the monetization engine—that almost undersells how hard it is to build one.

Mortgage—I’ve had to get fingerprinted in God knows how many states. We're doing this Mr. Cooper deal, and I had to send ink fingerprints to the state of Virginia, plus every bank account I've ever had. It's just so complicated. I wish it weren't that way, but it turns out it's very, very complicated to add on and bolt on a monetization engine. And, by the way, it's very hard to build one of these daily-active-use products.

Because you have a little bit more voyeurism in the field of real estate, it's not necessarily that I have very near-term proximity to purchase intent. Why hasn't Zillow done this? You know why? I'll tell you why: it's really, really hard, and they're addicted.

Yeah, I would just add that winning in housing is not for the faint of heart. I just don't expect that an early-stage company in a garage is going to be able to deploy. There's so much activation energy that's required. Regulation is very state-specific. Products are fragmented. Distribution is hyperlocal, and you have cyclicality where volume swings with rates. Lenders have to scale up and scale back.

You have banks stepping in after the financial crisis. You have the secondary market with originators, aggregators, and servicers. You have a bunch of legacy processes: manual docs and appraisals. You also have all kinds of other things, right? You have employers that have to verify income, banks that have to confirm your assets, and insurers, appraisers, title companies, and inspectors. Everyone uses different timelines and systems. It's just friction and fragmentation.

Part of it is that the amount of activation energy required to decide to do that is just not for the faint of heart. That's one of the things I love about what we do: we've decided to overcome that activation energy. And guess what? It took us 40 years.

Why is it that credit-card networks have such a dominant presence in the market—the big ones, Visa, Mastercard, and Amex—and why is it that you have a bunch of neobanks creating new credit cards but not taking significant share? The simple answer is that it took them a long, long time. It wasn't something that just happened, where there was virality and a network effect and suddenly these things grew.

Some of it is just the simple fact that the easy way is the long way, and the long way is the hard way. There is no easy way, and that's something that I think gives us a little bit of a first-mover advantage. It's not something that we take for granted, but the amount of serious fragmentation, disparate technology and systems, and the incentive models around the different players are all at odds with each other. We look at that as a good problem because we've overcome some of that activation energy.

Now you have technology and AI, automation, the ability to compress turn times, lower fixed costs, connect the different pieces, and build a more connected journey. That's where I think it's a good problem to have, but it's also one where we have an opportunity to do significantly more, given that we've overcome that activation energy.

That's a good note to wrap on. Guys, thank you so much for coming on the podcast.