Okay, great. Thanks, everyone, for joining. My name is Rajat Gupta, a member of the Automotive Equity Research team at JPMorgan. We're very pleased to have with us Mark Jenkins, Chief Financial Officer of Carvana. Mark has a few slides he would like to run through, and after that, we'll go to Q&A. Thanks, Mark.
Great. Thank you very much for having me. It's always nice to be back at this conference. I'm excited to be here today talking to you about both our medium-term plan and also our Q2 results, which I think are a nice step forward in executing our medium-term plan. The standard safe harbors apply to this presentation. Carvana is the leading platform for buying and selling used cars online, as many of you know. We recently put out a medium-term goal to sell 3 million cars annually in five to 10 years and to achieve a 13.5% adjusted EBITDA margin when doing so. Our Q2 results are a really nice step forward on executing that plan. It was a quarter of many records. We set a record for the number of retail cars that we sold in a quarter at just over 143,000. That was a company record.
It reflected growth of 41% year- over- year. We also set company records on many key profitability metrics, including adjusted EBITDA margin, net income margin, and various operating and net income dollar metrics as well, which I'll talk a little bit more about on future slides. We're marching forward. We're growing fast. We're doing so at very strong levels of profitability, which is all part of our plan. I think one of the things that we've been most excited about over the past year or so is our competitive positioning. The platform that we built, which is highly vertically integrated and has a customer offering that customers love, is leading to both industry-leading growth and industry-leading margins at the same time. We think that is historically unique.
We think that positions us extremely well for a long growth trajectory, particularly in light of the fact that where we stand today, we're only about a 1.5% market share in our core market of U.S. used vehicles. I think we have a long growth runway. We're achieving this level of growth of 41% in Q2 in an environment that is pretty stable. We have some data here on traditional auto retailers that grew about 1% same-store sales year- over- year. We're outgrowing the average of the public used retailers by about 40 points. I think we'll talk a little bit about some of the fundamental drivers that are driving that growth, of which we'll point to three that really feed back on each other. I think that's obviously a very strong performance and very strong growth profile that demonstrates how much customers love our offering.
At the same time, we're operating at adjusted EBITDA margins that are roughly 2x our peers. I think that's really stemming from the benefits of vertical integration. We built the Carvana platform from the ground up to be deeply vertically integrated. We have our own logistics and delivery network. We have manufacturing-style reconditioning that is integrated with that network. We built a website with an integrated financing platform to try to give customers a seamless experience. All of that vertical integration and all of the components of the platform that have been built together from the bottom up to provide this online experience are also leading to very strong margins. We hit a key milestone this quarter. We've been the most profitable auto retailer by adjusted EBITDA margin for several quarters in a row now.
This quarter, we hit a new milestone, which is now for the first time, we're the most profitable auto retailer by GAAP operating profit as well as net income dollars. For a long time, really in our entire time as a public company, we've had a goal to be the largest and most profitable auto retailer. We think we've definitively checked the second of those boxes. We're very clearly the most profitable auto retailer. We're also the fastest growing auto retailer. We're not yet the largest, but that's certainly our goal, to move from being the fastest growing and most profitable to the largest and most profitable simultaneously. What is driving this growth? We really talked a lot about the three fundamental drivers of our growth.
We really see these as key growth drivers for the past and how we've gotten to where we are today, selling now 143,000 units in Q2 at industry-leading margins. We also see them as key fundamental growth drivers for where we go from here. There are three that I'd point to. One is just continuously improving the customer experience. This means lots of things, but I think we had a nice example in our shareholder letter of a situation. It's a record for us where a customer came to Carvana.com, got a value for the vehicle that they wanted to sell, and then had dropped off that vehicle and gotten a check in just 38 minutes. That's a really fast and seamless transaction.
That transaction is an outlier, but it's the type of customer experience that we want to provide to customers and the type of customer experience that we believe is very, very differentiated and also one that requires very significant investments in systems and a fulfillment network to be able to make those sorts of customer experiences possible. Continuously improving the customer experience is a key driver of our growth and one we expect to benefit meaningfully from as we continue to move forward. The second key driver of our growth is increasing awareness, understanding, and trust. I think the simplest way to think about this is for people in this room, Carvana probably feels really familiar. The reality is we're in the very early days of e-commerce adoption in automotive retail.
If you look across the broader retail economy, e-commerce penetration is getting up just shy of 20% on that order, whereas automotive retail is down in the 2% range. We're really early in the e-commerce adoption curve in automotive retail compared to other retail verticals. As a result, we think there's still a lot of opportunity to raise awareness of our offering, explain to customers the benefits of our offering, and continue to build trust in the offering as a long-term fundamental growth driver. Last, in terms of the fundamental growth drivers that have got us here and we expect to drive us going forward, is inventory selection and other sources of positive feedback. Selection is a big growth driver. The used car market is a market with many, many skews when you look across year, make, model, trim, mileage, color, packages, and options, and so forth.
I think we're also very early in our path of having the broadest selection that we can have that drives positive feedback. The more selection you have, that means more inventory pools, which puts cars closer to customers, gets faster delivery times, and so on and so forth. These are the three fundamental growth drivers that we are executing against today. It's driving our 41% growth, and we're achieving that growth, like I said before, at industry-leading margins. The last component of our strategy that I wanted to talk about, it's one of the key things that we're focused on today, is really just expanding our network. In 2022, we acquired ADESA, a nationwide physical auction with 56 physical auction locations. We're in the process of integrating those locations with the Carvana retail platform.
One of the things that we've been focused on is site by site, going in and integrating ADESA physical auction locations, bringing in Carvana retail reconditioning processes, technology, including our proprietary Carly inventory management system, connecting it to our first-party logistics network that allows us to get cars from those integrated ADESA locations to customers as quickly as possible. We've been steadily doing that. That increases the number of inventory pools that we have available and puts cars closer to customers where we're acquiring them. One stat on that, by integrating more ADESA locations, we've actually seen the average inbound transport distance from cars we acquire to customers and then recondition drop by about 20% year- over- year. It also puts cars closer to customers for the purpose of selling cars to customers. On that note, we've seen outbound transport distances down about 10% year- over- year as well.
It also leads to faster delivery times to those customers. Integrating ADESA locations, adding more inventory pools, increasing efficiency of our fulfillment network, and giving better customer experiences, including faster delivery times, is core to our strategy today. That's something we've been marching out over the last 12 months or so and would expect to continue marching out in this early phase of our multi-year growth plan. That's a little bit of an overview of where we are, industry-leading growth, and industry-leading profitability, as well as the core element of our strategy. This last slide is just some additional highlights. I'm not going to go through each of these so we can move to Q&A. I think we're very strongly positioned with just a 1.5% market share, a strong nationwide infrastructure, and industry-leading growth and margins to continue to execute against our goals.
With that, I'll pause and we can take some questions.
Great. Awesome. Thanks for that quick summary. I wanted to start with just the fundamental gains topic. I believe it was last year in the third quarter when you had first hinted or talked about passing on fundamental gains to consumers in conjunction with the phase III entry. If you look at your last 12-month growth rates quarter by quarter, they have continued to accelerate, while at the same time, your margins or EBITDA per unit have also continued to expand. I guess the question is, have you already started to pass on these fundamental gains that you're seeing over the last 12 months? If you could highlight some areas where you have passed those on or you haven't really entered that phase yet.
Sure. Let me start by talking about some of the key areas where we see opportunities for further fundamental gains in unit economics. I think it's natural to think, OK, you already have industry-leading margins. Is there more room for you to continue to be more efficient? Is there more room for you to continue to drive additional sources of revenue on each car you sell? The answer that we really believe is absolutely yes. Wherever we look throughout the business, whether it's any gross profit line item or any expense line item, we still see meaningful opportunity for efficiencies in variable costs and revenues. We call those fundamental gains. Reducing variable costs, increasing variable revenues. We also see very significant opportunity for overhead leverage. As a data point on that front, our current physical infrastructure, we believe, supports annual sales of over a million units.
We're selling meaningfully less than that today, so there's real opportunity for overhead leverage as well. Most importantly, as much progress as we've made over the past several years on the economics of our business, we still see meaningful opportunity for fundamental gains, both in variable costs and revenues, and certainly in overhead leverage as well. Now on to the question of passing back gains to customers. That's something where we also see opportunities to do so over time. We've laid out this medium-term goal to sell 3 million units annually in five to 10 years and do so at a 13.5% EBITDA margin. From our perspective, you add up all the areas where we see opportunities for fundamental gains as well as the overhead leverage opportunities we have, we believe we'll have opportunities to pass fundamental gains on to customers.
That's something that we do want to do as we continue to make fundamental gains, and it's a place where we see a lot of opportunity.
is not something you've already meaningfully started to pass on to accelerate the growth that you're seeing right now.
I think that we don't expect to be overly specific on the breakdown of how many fundamental gains are going to margin and how many fundamental gains are going back to customers. I think I would say most of what I just described, I would see as future opportunities.
Understood. Maybe you're hitting on one of the points that was in the second quarter call and kind of tying to the advertising question a bit. We know that used car demand is highly seasonal, with peaks around tax refund season, slower periods later in the year. How do you manage staffing, reconditioning, logistics capacity to handle those peaks without overbuilding permanent infrastructure? When seasonality creates excess capacity, how do you decide whether to fulfill it through demand generation investments like advertising or just maintain steady operations to protect economics?
Sure. Yeah. There is seasonality in our business. The business has some seasonal peaks, and it has seasonal cadence over the course of the year. That seasonality is just like other used car retailers. It's not unique to us. I think when you're planning for seasonality, there are various approaches. If you have a seasonal surge, for example, you might plan in some flex capacity. Most of our operations have some degree of flex where you could shift hours up and down. You could drive your transportation fleet more or less. There are a number of places where our operations have flex capacity to handle things like seasonal or week-to-week variations. We take advantage of those. That's a tactical point.
Also, to your point, I do think there's some operational benefits from trying to smooth out seasonal fluctuations to some degree because it means you don't have to flex your operations quite as much. I think we have some tactical levers to manage seasonal fluctuations that are common in the industry, for sure.
Understood. In the same way around advertising, if you look at site traffic, it's back to levels we've seen at past peaks. Clearly, the top of funnel looks pretty healthy. As you think about taking the next step in growth, how do you balance putting more dollars into brand advertising to expand reach versus focusing on converting the traffic you already have more effectively? Given the long purchase cycles in used cars and auto retail, how are you evaluating the ROI and payback from these campaigns, especially when AI automation could just be lowering the cost of creative testing, personalization, et cetera?
Sure. Yeah. Lots in that question. One of our fundamental growth drivers and a place where we expect to invest over time is building awareness, understanding, and trust of our offering. A key way to do that is advertising. In particular, brand advertising is helpful in telling your story, telling customers about the strengths of your customer offering, what differentiates it from the other things that they're used to and that have been out there historically. We do expect to invest in advertising over time. I would expect a strong component of that to be brand advertising from the standpoint of telling that story.
I think we've actually seen really good results from our brand advertising in the past, just going in and looking at individual markets and seeing what happens when you do brand advertising in individual markets, how the sales grow over time, how your market share grows over time, and how your advertising costs per car sold actually decline over time as you stack brand advertising on top of brand advertising in markets. I think we have good data points that the long-term payoff for brand advertising is meaningful and expect to continue doing so. In terms of measurement, you made a couple of other points. I do think we're optimistic, the way that we look at it, that AI will be helpful in making brand advertising more efficient.
It just allows you to test more messages and more diversity of content more quickly at a lower cost, which I think can raise the overall efficiency of brand advertising. In terms of measurement, I think brand advertising is more challenging to measure than deep in-funnel performance marketing. I think we have tools to do so, including market or regional testing and other ways to actually try to get attribution from your brand advertising.
Understood. Maybe one question to supply before I open up to the audience for more questions. You've mentioned, I think, in the past that your current selection is still only a small fraction of the total used SKUs in the market. How do you prioritize which segments, trims, configurations to add to maximize the match rate? How does that expansion translate into higher conversion rates and faster delivery times? Relatedly, as you grow your breadth, how do you balance the benefits of just that wider selection with the complexity of stocking and just moving more of these unique SKUs?
Sure. Yeah. To hit on the beginning of your question, we absolutely see an opportunity to expand the selection of vehicles that we have on the site. That's a nearly uncapped growth lever over time. There are so many SKUs in the used market. Having more selection just makes it more likely that you're going to have the car that a site visitor is looking for. I think we absolutely plan to grow selection over time. I think growing selection over time can also have benefits for delivery time. If a customer is looking for a specific car, they're more likely to find that car in their market or one market over, as opposed to only being able to find that car or something that's close enough for their preference as many markets over. Definitely a long-term growth driver.
In terms of how we determine the right selection, it is highly algorithmic and highly data-driven. We're constantly looking at the demand that we're observing on the site. What vehicles are people searching for? What vehicles are they clicking into on the search page? Certainly, what vehicles are people moving down the funnel on and transacting on? We take all that information into account about what we're seeing on customer demand at a very granular level. I think the level of granularity that we use actually increases over time as our data grows. One of the places where we've been seeing fundamental gains over the last 12 months and would expect to see fundamental gains going forward is just making use of our growing data set. We've now bought and sold, I believe, over 6 million cars, if you count both retail and wholesale.
That data set is growing exponentially as our business grows. We're just constantly looking for ways to make more efficient use of that data in things like building the right inventory selection. That would be a few examples. It's certainly an opportunity for us over time to continue to grow selection and make use of our data to do so as efficiently as possible.
Understood. I'll just take one more before handing it over to the audience. Just on the lending side of things, if you look at your finance attach rate last quarter, just per our calculation, we think it was the highest ever. We've seen around 85%, 86%. You suggested that this increased attach rate has been driven more by just the integrated experience than just simply offering favorable APRs. How do you think about maintaining that penetration as you scale while also protecting underwriting quality and loan performance? Just compared to traditional indirect lenders, what do you think is unique about your selection and approval process that allows you to achieve both high volume and also favorable credit outcomes?
Sure. Yeah. I think from a customer experience perspective, having a vertically integrated finance platform is very valuable and very convenient. You can go on to our site, fill out a short credit application form, and our algorithms will then tell you exactly your financing terms down to the penny on every single car in our inventory. We have tens of thousands of cars. You know what your financing is going to be. It's very convenient, it's very transparent, and a lot of customers like it and choose to use it. I think having a vertically integrated finance platform is very efficient for the customer and very great for customer experience. It also allows us to streamline the process through the transaction because we're fully vertically integrated and we control everything. I think that's been a big driver of our finance attach rate over time.
If you look at our growth from being a very small company to where we are today with 143 retail units sold, for the most part, I think there's been some movements. Finance attach rate has gone up a little bit over time as we've made the product more convenient. Really, we've always offered that level of convenience, and we've always had a high finance attach rate. I think it's just the convenience drivers that I was just describing that has been driving that, certainly not just this year, but over the last 10-1 2 years.
Anything in just a specific underwriting capability you would say is differentiated, or just the approval process versus more traditional lenders that you've created?
Yeah, sure. I've talked a little bit about our vertically integrated financing as a driver of customer experience. I think it's also a driver of unit economics because we believe that a vertically integrated finance platform model is much more efficient than the traditional indirect lending model. If you think about the traditional indirect lending model, the customer goes to the dealer. The dealer is the party that has contact with the car. They're the one that has looked at it, inspected it. The lender hasn't necessarily seen the car, and the lender doesn't necessarily interact that much directly with the customer. Moreover, in the typical indirect model, you have the dealer sitting in the middle, the customer's here, and there's multiple lenders competing over here for the dealer's business and the customer's business. That can give rise to things like adverse selection.
If you're bidding at an auction against other lenders, that's an environment that can give rise to selection effects and things like that. I think our model, where we're fully vertically integrated, we know the car extremely well. We put it through a 150-point inspection in our manufacturing-style reconditioning centers. We have a 100-day limited warranty after the customer purchases the car to help ensure anything that comes up shortly after sale is also covered and the car is good quality at that phase. We really know the car well. We also really know the customer well. We're the only party that's interacting with the customer, so we can have very robust lending and underwriting processes there. Lastly, we're not participating in an auction like many indirect lenders have to do. I think those are some real fundamental advantages of vertical integration in this part of the business.
We have vertical integration in other parts of the business as well that yield similar types of benefits, in particular, a vertically integrated logistics network. I think vertical integration is a powerful tool for driving better customer experiences, which supports growth, and also driving better unit economics.
Understood. I wanted to just see if there were any questions in the audience. We went back there.
Thank you very much. That was very insightful. I wanted to touch on your point on variable costs. Would you offer some more color on how you're thinking about lowering variable costs? If there are any strategic or operational levers that you're thinking of, what would be a good milestone as you trend towards that?
Yes, absolutely. We've made significant progress lowering variable costs over the past couple of years. For those that are interested, we report a cost line item called Carvana Operations Expense in some of our financial tables attached to each quarterly earnings. That number has declined significantly over the past couple of years. The Carvana Operations Expense is about $1,550 per car today, of which around $300 is the 100-day limited warranty expense. Excluding warranty, all of the variable operations associated with fulfilling these sales cost us about $1,250. That's fulfillment, last-mile delivery, customer care centers, transaction processing, and any other operational expenses that are more variable in nature. That's very efficient. That's actually very efficient.
I think if you think about, for a near nationwide delivery network and executing a fully vertically integrated transaction, including financing, ancillary products, trade-ins, et cetera, we think that that's very efficient today and a big part of our success. Having said that, we still see meaningful opportunity, which is where your question was heading. If I talk about a few of the key sources of that opportunity, adding more inventory pools is one. I talked about as we integrate ADESA locations and add more inventory pools, that lowers inbound and outbound transport costs because we're just putting more cars closer to customers. AI is a huge opportunity in this area. I think we've made very significant gains on some of the more centralized tasks like customer care, various forms of communication, whether it's phone, chat, or email.
That's a big opportunity for AI where we've made gains but still see significant opportunities for further gains. I think things like document processing, title, and registration are very fruitful areas for AI to do an amazing job at having a streamlined, efficient, and very accurate transaction. Those are a couple of things that I would point to. There are also the business-as-usual elements of continuing to develop our technology systems that underlie logistics and last-mile delivery network, the reconditioning centers, the customer care centers, continuing to focus on operational excellence by having regular operating reviews and making sure that all the teams and operations that underlie those variable costs are executing as well as possible, always working to identify locations that are performing below the median and pull those to the median, identifying locations that are at the median and pull those to the upper decile.
Those sorts of operational efforts have yielded meaningful gains over the past couple of years. I would expect them to continue to yield gains over the coming years.
Any other questions? One here.
Sure. Yeah. Yeah. Our CapEx budget for 2025 is around $150 million. That is split across a number of different categories. A component of that is ADESA integrations. Just to be a little bit more specific about ADESA integrations, each site costs $2 million- $3 million. The reason that is fairly CapEx light is in this first stage of building out retail reconditioning capacity at ADESA locations, we're utilizing existing structures at ADESA. What it means to integrate at an ADESA location basically means take the existing structures, build it out with retail reconditioning equipment, install the Carvana proprietary inventory management system, and then implement Carvana-style retail reconditioning processes. It's a CapEx light integration. I also mentioned you connect it to the first-party Carvana logistics network. Over time, we would like to build out the ADESA locations much more fulsomely.
That will mean looking at the selection of 56 ADESA locations and actually constructing new structures on the plots of land where these ADESAs are housed. That will allow us to significantly further increase retail reconditioning capacity at those locations. The way we've sized that build-out in 2022, we sized a full build-out of ADESA at around $1 billion of capital expenditures. I think with inflation, it would be a bit more than that today, but it gives you a sense of the order of magnitude. We would expect that to play out over a period of multiple years, investing in that full ADESA build-out as we march toward our 3 million unit goal.
One more question here in the front.
As you're starting to work with a more, and I don't want to say mixed fleet, but that's the only word I can think of, of EV and ICE platforms, how does that change some aspects of the economics of the business just for how you evaluate reconditioning and make sure you're getting the right value for a car?
Sure. Yeah. I think the very simplest answer to that is we don't view EVs and ICE cars really differently from the standpoint of executing our business. There can be small differences in reconditioning, including what you recondition. There might be certain parts that tend to wear more on EVs and certain parts that exist and tend to wear more on ICE cars. It can change your mix of exactly what you're reconditioning. Overall, I would say we don't view them very differently. I guess another thing that is different is EVs require you want to have charging infrastructure, certainly at your inspection and reconditioning centers, but also at things like your last-mile delivery locations or vending machines so that you can make sure that cars are charged up to standard before you deliver them. Things like charging also involve a set of operational processes.
What are the standards for charging? How frequently do you circle back to a car to ensure that it has the right level of charge? There are a number of operational processes. We've implemented most of those. We're selling a lot of EVs today. I think our EV market share is noticeably higher than our ICE market share. We're having great success selling EVs. I gave you a few examples of the little differences. Overall, those are little differences between selling ICE cars and EV cars. We view them much more the same, more than we view them as having meaningful differences.
Got it. We have like a minute left. I wanted to make sure I ask a standard CFO question.
Oh.
You've said you wanted to operate with investment-grade credit metrics long term, especially after the lessons of 2022, 2023. As you get closer to that rating, how should we think about the sequencing of deleveraging, refinancing some of this high coupon debt, maybe more investment in like M&A, like the franchise acquisition type stuff, and just eventually maybe returning capital?
Sure. Yeah. I think that we're generating now significant amounts of adjusted EBITDA. Our adjusted EBITDA is very high quality, so a large portion of that is cash. There's a question of, OK, where do you invest the cash? I think priority number one is investing in the business. We're integrating ADESA locations, we want to have more selection. There's a number of places where we want to invest in the business, and I'd say that's priority number one. I think a second priority is we do have goals to deleverage over time. We want to do most of that deleveraging through EBITDA growth, but we have repurchased some of our senior secured notes over the last year or so, and that has been a use of some of the cash that we've generated from operations. Those would be a couple of things that I would point to.
I do think over time, a lot of refinancing opportunities don't necessarily take cash from operations, but they do improve your capital structure if you have a lower interest rate as your credit metrics improve. I think that's another opportunity. For today, in terms of using cash generated by the business, I would point to investment in the business and then, to a lesser extent, selective opportunities to deleverage.
Understood. Maybe we can take one more question now. Good.
I think some of the key things that I pointed to, deep knowledge of the car and ensuring that it's a very high-quality car, a firsthand knowledge and interaction with the customer versus going through a third party to interact with the customer, and avoiding an auction or adverse selection. I think those things are difficult to quantify. I think that we can see them in our results where we have very strong and fairly stable other GPU, which is our F&I GPU. This is how other dealers might refer to it. We generate a lot of F&I GPU. I think that that's where you can see some of those effects flow through into our financial results.
Understood. Great. I think that's all the time we have. Thanks, Mark, for joining us.
Thanks a lot, Rajat.