Carvana vs CarMax: Comparing Operating Models

Disclaimer: This interview is for informational purposes only and should not be relied upon as a basis for investment decisions. In Practise is an independent publisher and all opinions expressed by guests are solely their own opinions and do not reflect the opinion of In Practise.

I've been researching the used car market and I believe there's a lot I can learn from you. I'm trying to understand all the stages involved in acquiring and selling a vehicle. I'm interested in how these stages differ or are similar for an independent dealer, a franchise dealer, and larger entities like CarMax and Carvana.

Essentially, I'm trying to understand how the economic value chain changes across these different footprints. I'm particularly interested in your perspective given your 20-year tenure at CarMax, which is the largest independent dealer.

Used cars are somewhat of a commodity, trading within relatively tight price ranges. For example, the price difference for an average $25,000 car might be around $1,000 to $1,500.

In theory, a business can thrive in a commodity-like sector if it is the low-cost producer of that commodity. A large-scale operation like CarMax should have certain advantages over smaller dealerships. These advantages could be leveraged in various ways, such as improving margins and profitability per unit, enhancing quality to capture market share, or using the excess value created from larger scale to acquire more, sell for less, and produce better cars.

When you have commodity-like pricing set by the market and non-commodity-like production costs, it opens up interesting opportunities for companies like CarMax or Carvana. I'm trying to understand if this hypothesis holds true.

The used car industry involves many stages. For instance, buying from a customer is different from buying from an auction. There are also considerations like moving and reconditioning the car. I'm hoping you can help me understand these different stages and how they might vary between a small player and a larger one.

Yes, I believe you're on the right track. The chain and its links you described are essentially the same. However, the difference lies in the two pipelines from which you're acquiring cars. You and I, as individuals, are often referred to as 'private party' in the industry. These are individuals not associated with any entity or company reselling in bulk. They are simply selling one or two cars, perhaps trading them in for replacements. The acquisition process is pretty much the same as you described. Typically, the car is bought in the local market and reconditioned at that location. Unless, as in CarMax's or Carvana's models, it's dropped off at a car tower or at one of CarMax's satellite stores. A good example is Los Angeles, where you have about eight satellite locations, several production locations, and a massive reconditioning center in Palmdale. This large facility is where cars from the LA Basin are trucked to for reconditioning before being transported back. This is more cost-effective than adding the labor cost and other associated expenses of reconditioning at the original locations.

How large is the facility in Palmdale?

It's enormous. When I left, they were reconditioning 1,000 cars a week. To give you a sense of scale, the largest facility in the LA market at the time was probably reconditioning 250 cars in a single store per week. So, it's four times larger. It's a massive facility that runs an auction. It's primarily staffed by technicians, detailers, painters, and a few associates to move cars around. CarMax has built a few more of these facilities. There was a time when they considered regionalizing this model across the country to maximize the footprint. The game-changer in this model is the auction. However, what people often overlook or are unaware of is the prevalence of auctions. For instance, Manheim has 100 auctions in the US, and ADESA has 70. These two major players are present in almost every major metro area. So, if I'm buying cars for the LA Basin, I can purchase them at an auction. Between LA and San Diego, there are probably three Manheim's and at least two ADESA's. This means that for the LA Basin, there are five locations, minimizing transportation costs as you're only moving the cars within that area. It's not like you're transporting cars from LA to Dallas, for example.

This might be jumping ahead, but I'm curious because I visited an ADESA location in Southern California, specifically in Riverside County. I got to see the live auction on a Friday morning, which was quite fascinating.

It's quite an experience, isn't it? Auctions can seem a bit chaotic, right?

There's a significant variance between buying and selling parties due to the possibility of pre-auction reconditioning. For instance, if you're CarMax, do you purchase everything in a non-reconditioned state? This is because you have a large facility where you can conduct the reconditioning process, which is likely to be cheaper using your own labor and parts. This is assuming that local transportation isn't costly. But if you're doing as much volume as CarMax, transportation costs shouldn't be too high.

Yes, it's typically a short haul.

You could use a nine-car hauler, for example. However, the sellers have the discretion to recondition the cars before the auction.

Indeed, and the buyers can also pay to recondition the cars post-sale. Both ADESA and Manheim offer this frontline ready process to either the seller or the buyer.

CarMax, however, will never purchase a frontline ready car. This decision is based on four variables: cost, quality, speed and safety. When they factor in safety, it’s not just about the safety for their employees to do the work, it's the safety for their customer. They won't waiver on safety, and that's why they would never allow it unless CarMax worked a deal with them and was co-located on those properties. And it was CarMax employees following CarMax processes, using CarMax resources to recondition those cars.

Have they ever done that?

No, CarMax hasn't done that. However, Carvana and Vroom did that with Manheim years ago. Vroom, originally known as Texas Direct in Houston, had only one reconditioning center when it started. They then made a deal with Manheim to co-locate reconditioning at several locations. This strategy is typically adopted by smaller companies that lack the scale and financial resources to support the reconditioning process. There's a lot happening at these auctions. For instance, at Riverside ADESA, there are likely 30 different lanes with cars running and auctioneers calling out bids. It's quite a spectacle.

And the cars are running through at about 35 seconds per car, correct?

Yes, the cars are moving quickly.

I noticed they had a Tesla-only lane that was even faster, at about 20 seconds per car. It seems like the larger buyers do most of their inspection work ahead of time.

Indeed, CarMax is usually there a day or two ahead of time, previewing cars. However, consider this. It's quite amusing to see everyone with their caps pulled down low and sunglasses on. It's reminiscent of the World Series of Poker. The most interesting part is that everyone knows each other there, including the auctioneers and the people selling the cars. They know who you work for.

So, the moment a CarMax buyer steps into any of those lanes, the seller knows they have a golden opportunity. They know that the buyer is there for a reason, that they have previewed their cars. They know there are cars in that run that they want to buy.

So, the moment that hand goes up for CarMax is when the seller gives the auction block the green light to accept the bid on the car, even if it's below what they started at. This is because they know they can sell many cars to CarMax.

I want to ensure I understand this. If I'm watching the live auctions and CarMax is bidding, let's say the car is at $19,000. Are you saying they'll sell it to CarMax for $18,500 if that's their bid?

Yes, that's correct.

They'll accept the bid immediately if it's from CarMax?

Yes, they will. If CarMax places a bid, they'll close the deal immediately.

They won't even wait for anyone else to respond?

The moment we raise our hand, the deal is closed.

Because CarMax is a preferred buyer?

Yes, because they know that CarMax is not only buying that car but also another 99 that day from that auction. The auctioneers and sellers are all aware of this.

Now, it wasn't always this way. About 20-25 years ago, we had to assert ourselves more. We were big spenders and we weren't always polite about it. We would come in and they knew we were going to spend a couple of million dollars that day. On the other hand, a dealer with only one location and $30,000 to buy two cars isn't buying many cars.

So, there's a scale function because they know CarMax is going to buy a lot of cars. I'm not talking five or ten or 20, I'm talking hundreds at most of these auctions. Years ago, the split was probably 75-25, with 75% of the cars sourced through the appraisal lane and the rest through auction. That ratio flipped probably eight or nine years ago when individuals started holding on to their cars longer.

When you say appraisal lane, are you referring to a trade-in from an individual?

Yes, that's correct. 20 to 25 years ago, we didn't acquire as many cars from auctions as we do today. The increase is due to more people using auctions to dispose of their vehicles.

Additionally, many fleets and rental car companies use auctions to defleet. Often, they have a set price in mind for each unit. If they don't achieve that price, they'll approach CarMax before the auction ends. Even after the auction, people are still there settling their bills of sale, making payments, and arranging transportation. Rental car companies will seek out CarMax buyers and offer them unsold vehicles at their desired price. CarMax often accepts these offers, viewing them as opportunities to acquire a discounted lot of cars.

There are also pre-negotiated deals. For instance, when Hertz launched their Teslas, they had a backend agreement with Carvana to defleet the Teslas directly to Carvana for sale. Such agreements are common.

Due to their scale and buying power, CarMax and Carvana command a lot of respect at auctions. They are given a wide berth by both the auction house and the sellers. The only ones who aren't fond of them are the independent buyers.

Interestingly, despite being a major player, CarMax only represents about 2% of the used car market. They have a larger share of the non-peer-to-peer or private party market.

If we exclude private party sales from the equation, CarMax's market share significantly increases. However, when private party transactions, like individuals selling their cars on various platforms, are factored in, everyone's share decreases due to the large volume of such transactions. This is the mystery that everyone is trying to solve.

Peer-to-peer transactions, yes.

Exactly, how can they tap into that market? That's going to be the challenge.

Before we move on, I'd like to clarify something. If a large-scale company like Carvana were to fill up its capacity, what kind of advantage do you think they would have over independent buyers at auctions? They bring millions of dollars to an auction and can buy unsold vehicles. They also get first choice. Would the advantage be around $200, $500, or $100 per car?

I would estimate the advantage to be around $500 per car.

If I'm a CarMax buyer and you're a Carvana buyer in the auction lane, and there are 20 independent lot buyers, the seller is only looking at us.

So, it's just us, right?

I've attended these auctions and observed how some bidders are ignored by the auctioneers. They allow the bidding war to be dominated by the two powerhouses. It's quite interesting, almost like a game of cat and mouse.

Are there any other players in this club? Who else is part of it?

EchoPark is starting to gain some traction. That's Sonic's used car venture. Shift was there too, but their model wasn't sustainable and they imploded. Even Shift, despite their size...

They had around 20,000 units, I believe. That's perhaps large compared to independent businesses.

But Shift was still relatively small. They recruited a number of former CarMax employees. They had some leadership on the operational side and even brought in some CarMax folks. They were solid people, but they couldn't make it work because they didn't have the foundational structure for the business.

They needed the infrastructure.

Yes, they were top-heavy and it just didn't work out.

Let's shift gears to another question about acquisitions, if you don't mind. We've discussed the auction side, but I'd like to talk about the appraisal lane side and the quality of the inventory sourced from the auction lanes versus the appraisal lanes. How should I perceive this? For instance, Carvana, I believe, sources about 80% of their cars from customers, while CarMax sources about 75%, if my calculations are correct. This shift from 20% to 70% caused a lot of operational difficulties in reconditioning for Carvana, as it wasn't as consistent in terms of the types of units, their age, and mileage. It was generally more complicated. However, I'm not sure if you think it's worth it because when you're buying from a customer, there's less price transparency. They're not in the auction lane hearing the prices. They can check the Kelley Blue Book, but they don't know exactly where their car should be priced within that range. What are your thoughts on acquiring from customers and how does that look?

I've always believed in the old trade-in model. Typically, it's accompanied by a purchase. The innovative aspect is that they've created an option where you don't have to buy a car, and you can sell your car. They've used this opportunity to try to attract customers away from competitors. There's a small group of customers who sell their cars because they've acquired them through an estate, or as a gift, and they don't want to take on the burden or the taxes.

Regarding the numbers of 75% and 80%, I believe there's some ambiguity. If I sell 20 cars today, my goal is to make a purchase or a trade on 50% of those. But out of those ten cars, I'll probably designate a third for wholesale immediately because they don't meet my parameters due to high mileage, mechanical limits, or cosmetic issues. So, a third of those cars are going wholesale.

Let's say three of them go wholesale right away, leaving seven in the pool. That's the only way they can claim that 75% of our cars from customers go to reconditioning. What they don't tell you is that out of those seven, probably half of them get rejected.

They get moved from reconditioning to wholesale?

Exactly. They enter the reconditioning process and if something is found, or the reconditioning estimate exceeds the cost parameters, they get rejected. So, after clearing the ambiguity, I'd say it's more like 35%, or 30% to 40%, if you want to give it a range. If I bought ten cars at a CarMax today, on a good day, only four of those would make it through my shop. The others would either be designated for wholesale or get rejected.

I'm more familiar with the Carvana numbers because I've been looking at them more closely. But I think what's happening at Carvana is that if they sell 100,000 cars, they buy 130,000, and then they wholesale about 50,000 of them. So that means they're buying another 20 from auctions or other dealers. They're actually buying more than they're selling and then filling the gap on the retail sale number with wholesale units.

One aspect to consider is how Carvana represents their wholesale sales. When Carvana acquired ADESA, many thought it was a brilliant move. However, this acquisition led to CarMax and others deciding not to shop at ADESA, opting to shop at Manheim and other regional players instead. They didn't want to give their money to a competitor. This is why CarMax has never bought an auction. They understand that while they would gain a great source of cars, they would also lose customers buying from that source, at least in the short term. Some customers have started to return, but I would be surprised if CarMax actively starts buying a lot of cars from ADESA again.

Another thing to remember about auctions, and this also applies to individual car trades, is that about 90% of our trade-in activity is not because we love the car. It's because there's an issue with the car. This could be mechanical, cosmetic, or related to a previous accident. We may not like the way it was repaired, or we didn't get enough to repair it, or we simply don't want to repair it.

Are you referring to the wholesale activity or the trading activity?

I'm referring to trades as a whole. Cars that come in are either going to be wholesaled or retailed. Customers don't get rid of cars because they're driving great and they love them. They get rid of cars because they're no longer satisfied with them for a variety of reasons. You're not buying someone's treasure, you're buying their problem. Now, you have to decide whether to designate this car for wholesale or retail.

Sometimes, we buy a car not because we want it, but because it's part of acquiring a sale. We buy the car from the customer, finance them for a new car, sell them a gap, an extended service plan, and create a lifecycle with them. At other times, we may decide to designate a car for retail if it's in good condition. If it passes through the shop with a clean bill of health and low recon costs, we've created another sale off of your trade-in. So effectively, we've created two transactions with you.

They're profiting on both sides?

Yes, two sales through a single purchase. However, if a car comes back from the shop with high recon costs, we have to consider if wholesaling it would result in a loss.

If the car gets rejected, would it be a loss in the wholesale market?

Typically, if I purchase a car for 10,000, it will cost me 7,000 to recondition it. This means I've invested 17,000 with no additional fees. I intended to retail it for 16,900, but now I've spent 100 more than my intended retail price. The wholesale value of the car is 11,500. Let's use these numbers as an example. Now, I could decide to wholesale it, but if there are major engine issues or the car is burning oil, I have to announce that. So, I won't get 11,500 for it. I'll likely get 9,000. I'm going to lose 1,000 in fees, so let's call it 1,500 in wholesale. But if I invest 7,000 in it and add another 500 in fees, I'm in it for 17,500. However, I'm unlikely to sell it for more than 16,900. In this case, I only lose 600 on that car.

You take a smaller loss.

Correct. That's the calculation used in the used car industry traditionally.

I visited a Carvana reconditioning center in Tucson. I wanted to see how these centers operate. It was clear to me, going through the different stages of inspection and reconditioning, that they are running this decision tree analysis at every stage. They are constantly assessing when a car should be rejected, how many parts are too many, how many panels are too many to paint, and so on. It seems like they're eager to reject a car if it's not profitable for them, or they're simply optimizing for the margins on that car. I'm curious about how robust these processes are at a place like CarMax, where you've dealt with millions of units. It must be quite robust, but there might also be some judgment involved. I'm interested in knowing if there's a lot of judgment involved or if it's largely formulaic.

Our process is very formula-driven. In fact, during 2008, 2009, and 2010, when we were focusing on our standards, we used a phrase that still resonates today is "Get off your gut and get on the guide. You would be amazed."

I was part of an initiative where I visited various stores and worked with employees who had been with the company for 10 to 15 years. They were planning extensive work based on cosmetic inspections, but I would often intervene, suggesting less costly alternatives like wet sanding, buffing, and touching up instead of painting multiple panels or replacing wheels.

We have comprehensive standards for both cosmetic and mechanical aspects. These standards take into account the price of the vehicle, its drive type, and other factors like tires and brakes. Every aspect has an expectation. Carvana follows a similar approach. It's rare for someone to make a decision based on instinct.

These guides were introduced to eliminate gut decisions, which were common in the early days at CarMax. These instinctive decisions often led to conflicts between purchasing, operations, and sales departments. For instance, the purchasing department would buy a car at a great price, but then the operations department would want to invest heavily in reconditioning it, which would then affect the sales department's ability to sell the car at a reasonable price.

We managed to eliminate these conflicts by implementing these standards. Now, we have written decision trees, either in electronic or printed form, that guide every decision, from whether to paint a panel to whether to replace a tire. Over the last five or ten years, we have expanded and slightly loosened these standards, but they remain very specific to help us make margin decisions.

I find this approach fascinating. My conclusion is that companies like CarMax and Carvana, which have systematized and made their processes more formulaic, probably do this well at scale. I believe that when processes are standardized and driven by a strict procedure, the variance per unit is tighter. This likely results in a higher quality average car coming out of these places compared to those that don't follow standardized processes. Would you say this is accurate, or is it possible that they standardize to a lower standard?

CarMax has refined its approach significantly. They rely heavily on focus groups and survey data from sales, service, and appraisal customers. They thoroughly analyze this feedback and use it for comparison.

Here's a metric for you to consider. When I first joined CarMax, it was quite easy to adjust your all-in cost on a weekly or even monthly basis. For instance, if your all-in cost for the previous week was $1,000 and someone from corporate asked you to reduce it to $800 per car, it was feasible. You could simply instruct your shop foreman or cosmetic manager to cut $100 from each car's cost, reducing the detail time from four hours to two. This would effectively remove two hours of labor from every car.

You could manipulate the costs as you wished. However, from 2010 onwards at CarMax, this was no longer possible. The only way to achieve this was to scrutinize every detail. For example, we had to question whether every car needed new wiper blades. Some locations, like the one you visited in Arizona, would insist on replacing the wiper blades on every car due to the desert heat. But considering the infrequent rain, was this really necessary? This is an example of the shift from relying on gut instinct to following a guide.

That's a standard change, though. It's not a judgment call, right? You're changing the guide, not the gut.

Precisely. The challenge was getting people to understand this. On a larger scale, you could consider changing a standard.

For instance, within the last five years, CarMax changed its policy on trucks with lift kits. Previously, even if a lift kit was professionally installed and had the correct tires, CarMax would not sell a vehicle with a modified suspension. This changed five years ago. The change actually reduced costs as we no longer had to remove and replace these components. It also allowed us to competitively price these vehicles, as they were desirable to customers.

The same applied to any custom features on a car. CarMax used to insist on factory standard only, but this added unnecessary costs. As I was leaving, we were working on a program to determine what custom features we should leave on cars based on customer preferences. We conducted focus groups to identify the most common modifications customers made to their trucks, which turned out to be leveling or lift kits, wheels and tires, and exhaust systems.

Once we learned this, we stopped removing these features, and it made a significant difference.

Indeed, it can be a positive attribute that people seek. Let me pose a high-level question. If you possess ownership economics and scale, you can presumably purchase cars at a lower cost than smaller businesses. Your reconditioning process is likely more efficient due to large, centralized facilities with standardized procedures. You're likely processing cars more consistently, your labor is more efficient, and you're probably purchasing parts at a lower cost due to procurement scale and bargaining power. Therefore, your costs per unit are likely lower than a small business doing reconditioning.

If you have a $500 advantage on acquisition and perhaps another $500 on reconditioning, you have a $1,000 advantage just through the reconditioning process. It seems that everyone, including franchise dealers, CarMax, Carvana, makes around $2,000, give or take, of metal margin on cars. Do smaller businesses make $1,000 on metal margin? They likely make up the difference with financing and other services like gap waiver insurance and extended warranties.

Are they competing with a lower margin? Or do they say, "I'm at a $1,000 disadvantage, so I'll recondition this car for $500 instead of $1,500 or $1,000 instead of $2,000," and do less work on the cars? Are they sacrificing quality to maintain the same metal margins as everyone else to make their business viable?

This may sound a bit sarcastic, but the answer is yes to all of it. They will adjust based on what they need to do to keep their dealership running. If you think of a small business, not a franchise dealer, they will find other ways to make the margin.

They also have car sales, trade-ins, and other revenue streams.

At a franchise dealership, the primary source of income isn't car sales. Instead, they make their money from the service provided after the sale in the Finance and Insurance (F&I) department.

However, if I'm running a small-scale used car lot, with about 15 cars, my operations would be different. I might have my brother-in-law working on the cars, his child washing the cars, and my neighbor's son driving cars to refuel.

My customer base is also different. These are customers who likely didn't get approved at CarMax or xCarvana or at the franchise dealer. So, when they come to me, I have the flexibility to charge them as I please or offer them a great deal. Small independents like me typically price the car just above the market rate and then allow it to come down. They're not a fixed price option and are willing to negotiate.

However, these small-scale dealerships also cut corners. They don't always fully recondition the cars. Usually, they buy cars ready for the frontline from the auction. If they buy a car from an individual, they don't follow a checklist, but rather what the brother-in-law remembers. Their primary goal is to keep the business running month to month, just like the payments they receive from their customers.

I'm not sure if this information is publicly available. I've tried to find it but haven't found any data. If it's proprietary, please don't share. But, do you have an idea of the reconditioning capacity that CarMax has?

I don't think they publicly disclose it, but you can estimate it. If you consider their sales, they have to recondition every car they sell. If they have that capacity, and if they buy 75% of the cars from customers, you can start to estimate their capacity.

For instance, if they buy ten cars a day, three of those are automatically going to wholesale. So, their capacity is likely around 20% more than what they sell.

It's probably a little over a million plus, considering they sold around 900,000 units in fiscal 2022. So, they might be handling around 1.2 to 1.5 million units.

They have a split model for paying their technicians and detailers on the production or reconditioning side. Half of their locations, or perhaps a bit more, operate on a fixed income or fixed pay model. Here, employees work 40-hour weeks on a fixed hourly pay with a salary target.

The other half of their locations operate on a traditional model where employees are paid on a flag rate basis, meaning they get paid per job. The advantage of this model is that technicians will aim to flag more hours than they clock. For instance, if they clock 40 hours, they would aim to flag close to 60. This makes them 150% efficient. If they're making $30 an hour and get paid for 60 hours, it's significantly better than getting paid for 40 hours. This mindset is common among technicians, detailers, and painters in the automotive world.

In locations where employees are still on flag rate, they can increase their workload faster and sustain it longer. This is referred to as the traditional model. The other model is called the flow model, where everyone is on an hourly rate and targets 40 hours of work.

Consider California, for instance, where the incentives are unusual. If you work more than 8 hours in a day, that's considered overtime. This isn't the case in Texas, where it's calculated per week. So, they have to balance that. They do a lot of what they call load balancing, which refers to the load of vehicles being put into the system.

When load balancing, they look at an area of the country and determine where they can increase product output. For example, if they can't increase output in California but can in Arizona, they'll increase productivity by 25% in Arizona, build the cars there, and then ship them to California to meet the demand. They use strategies like this to balance their production demand.

I enjoy learning about these nuances of the business. However, I'm wondering why, if they have a cost advantage in a fragmented market and are already a big player, they don't maximize their capacity. I understand there are affordability issues and macro headwinds on the sector, but why doesn't CarMax, as the largest player with the most scale and biggest cost advantages, capitalize on this while everyone else is under pressure? Why aren't they aggressively expanding their business and doing one and a half million units or whatever their number is, instead of being limited to around 800,000 to 900,000 units at this point?

During my tenure, I often asked about mergers and acquisitions type activity and why we didn't engage in it. The response I consistently received was that our employees are the most important part of our company. CarMax, at its core, prioritizes its associates first, customers next, and shareholders last. The belief is that if you take care of your people, they will take care of the customers, which in turn drives revenue and benefits the shareholders. The focus was always on creating an environment where people want to come to work and love their job. If you examine their annual or quarterly reports, or the accolades they've received, you'll see that everything revolves around this culture they've created. It may sound cliché, but that's the reality. They are willing to lose sales to maintain a culture that doesn't necessitate a union or create feelings of unfair treatment. This focus has become a cost for them over the years because they offer a comprehensive benefits package, including a pension, which is not offered by competitors or even considered by smaller, independent businesses.

Is it more of a labor staffing issue that they're not able to produce the units because it would impact the company culture to be at full production? Or is it that they are buying all the units they want and that's all they can get right now?

I believe it's a combination of the two things you mentioned. Firstly, it's a labor issue in that they don't want to hire people just for the sake of producing cars that they don't necessarily need or want to sell. As strange as it may sound, they don't want to be known as a "stack them high and let them fly" car dealer. They aim to provide a different experience for the customer. There are consumers who don't like buying a car from CarMax and will never do so because they prefer the traditional dealership experience.

They want to feel like they've won something in a negotiation.

Exactly, they want to feel like they've saved $500.

They really appreciate the hand-to-hand combat. Absolutely.

Many multi-rooftop groups in the dealership franchise world have adopted some form of the CarMax model because they've realized it's more effective. It fosters customer loyalty and recurring sales, and customers return for service. Regardless of who the CEO of CarMax was, the mantra was always the same thing, sometimes no sale is better than a bad sale. You don't want to just sell a car just to sell a car. You want to sell a CarMax car because it's a CarMax car, so that the customer has a CarMax experience and it mirrors and fits through that whole model.

This also applies to employees. Over the years, I've hired people who, it turned out, weren't a good fit for CarMax. It's a different world. I sometimes miss the incredible structure, the mentoring, coaching, and development opportunities, and the independence it offered. Unfortunately, many franchise dealerships are the exact opposite.

CarMax is about more than selling cars. It's about building a sustainable business with people who believe in what they're doing. This belief cascades down to the consumer, driving loyalty. Once a customer buys a CarMax car, it's difficult for them to buy from elsewhere because they appreciate the experience.

They're not just chasing volume for the sake of it. They want to do it in a methodical, thoughtful, and process-driven way.

Exactly, it's like eating an elephant one bite at a time. I'd rather have 10% of the share and do it the best possible way than have 20% of the share and only do half of it well.

That makes sense. Happy employees lead to better retention and a growing knowledge base. High turnover results in frictional loss, which affects the customer experience. People buy many cars in their lifetime, and if you can satisfy them the first time, they become lifetime customers. Over the lifetime of a customer, the most profitable thing you can do is treat everyone in the value chain, from employee to customer, in a respectful and satisfying way.

We often draw comparisons to the Nordstrom experience. When people hear "Nordstrom", they often think it's super expensive. However, that's not the case. They sell a wide range of clothing items. It's all about the experience, and some people may feel uncomfortable because they perceive it as being too expensive.

We also discussed how to create a customer experience similar to Chick-fil-A, where customers are hooked and love the experience. We trained our staff to use certain buzzwords in their communication and present options to the customers, giving them the power to make decisions.

We used an acronym, AIR, to ensure customers were Always Included and felt Relieved, without feeling pressured. The goal was to remove any anxiety from the car buying or servicing process. You might hear phrases like "I need to give my customer some air" or "I need to give that employee some air".

It was amazing because it created a unique culture. Having left them, I can say it's different from any other business.

Very interesting. Now, returning to my original question about the different links in the transaction chain, including acquisition, auction, reconditioning, and transport. When running a hub and spoke model with network infrastructure and production infrastructure, there's long haul and short haul transport. Smaller businesses may not have these, but they may pay more for the front end. How do you aggregate the cost of moving cars around and getting them to the right locations?

When they go to an auction, they aim to buy for a certain radius.

So they're buying in LA for LA, for example?

Generally speaking, yes. However, if Hertz offers them a hundred cars at a good price, they might buy and ship them, even if it adds a couple hundred dollars to the cost. They'll consider the western group of states as their options.

They'll also consider title transfer. How easy is it to move a car from California to Nevada, Arizona, or Oregon, compared to moving it from California to Illinois? If a car is not available for transfer on the CarMax website, it's usually because of title issues that could cause a significant delay in transferring the car to your name.

May I interject with a question that just occurred to me? I wanted to ask before I forget. This is about comparing Carvana to CarMax. It seems like CarMax has local regional production around all the major hubs where they have a presence. Carvana's production footprint, until they acquired ADESA, was more like an agricultural footprint. They produced in the middle of the country and perhaps had to acquire from the coast, then ship them in and back out. As an operations person, what was your impression of the Carvana footprint?

I thought it was inefficient.

And now, considering any competitive impacts with the auction business, do you think their current footprint solves many of their previous problems?

It solves a significant number of their problems. It gives them the ability to alleviate many pressures and move cars much cheaper than before.

Where do you think those pressures were most intense? Obviously, you weren't on the inside, but based on what you know and what you saw of their network.

The most difficult decision when moving a car from point A to B is whether to pass the cost on to the consumer. The only appropriate time to do this is when the customer requests a specific car to be transferred, for example, from Los Angeles to Houston, Texas. That's the only time you should expect the customer to bear the transfer cost. However, you can't charge an exorbitant amount either. You have to be reasonable so you don't undercut yourself, but you also don't want to push the price too high. So, you're not making a lot of profit there. That's not a revenue generator or a margin center for you. It's just a customer acquisition piece. In CarMax's model, that's a done deal. In Carvana's model, they incurred costs to move the car from an auction, let's say in Oklahoma City, to another location. They had to move that car from Oklahoma to Dallas, or to Tennessee, or wherever the closest Inspection Center is. Now with ADESA, they don't have to move it as far. Their initial transportation will be much more convenient and save them a lot of money. The problem they're going to have, though, and this always puzzled me, is why they didn't build their reconditioning centers in large metropolitan areas where they have a presence?

Why don't they have one in California at all?

Well, California is not attractive to anyone in the automotive business due to the stringent overtime rules and labor laws. Those labor laws are burdensome because they significantly erode profits. If you think about it, a reconditioning environment is no different than production. It's basically the same concept as building a car at a GM or a Ford plant.

If you can only schedule someone for an eight-hour day, they're technically scheduled for nine hours, considering breaks. They usually work around seven and a half to seven and three quarters hours. The lost productivity is significant. In states without these overtime laws, you can schedule people for ten-hour shifts, or even twelve-hour shifts. Many states don't have these daily overtime laws, only weekly ones. That's a significant factor.

Another issue is the high cost of moving cars into California. For some reason, it's the most expensive place to ship a car into and move a car around. Moving a car from Sacramento to LA is more expensive than moving a car from Texas to Arizona. It's just the freight lanes; they're expensive. The model for CarMax today is much different. However, Carvana still largely relies on other companies for their transportation needs.

As they build out the ADESA facilities, they've got a group focused on integrating the ADESA operation with Carvana to determine how to build the production lines.

When I visited Riverside, or as they call it, ADESA LA, they told me that they were planning to implement the Carvana IRC lines within the next few months. Currently, it's a person walking around the car, inspecting it, and taking pictures of imperfections with a smartphone. It's not as standardized as the Carvana way at the larger centers.

I believe that this will help them scale up and improve. They can also gradually expand the reconditioning capacity of the ADESA facilities. When they acquired them, they gained around 200,000 units of reconditioning capacity per year, which certainly helps. Having localized reconditioning eliminates a lot of the long haul, which I'm sure aids in network load balancing.

Once they get it rolling, they could potentially benefit from it. It might be the game-changer they need. I believe Carvana is at a point where they need a change in leadership. It's challenging when you're a founder; there's only so far you can take it. In my opinion, they've run out of runway. Having been in the business for a long time and having seen similar transitions at CarMax when Austin Ligon, the original CEO, exited, it feels like the same scenario. We were at the end of the runway and needed something different to progress further. That's how I perceive it when I read their reports, earnings reports, or any articles about them. It seems to me that either Ernie [Garcia] or his team needs to change. They need a different vision to take what they've built and push it further. The phase they're in now with this ADESA scale needs to be executed perfectly. If they mess this up, it could cost them heavily and potentially be disastrous. They don't need that right now. They need to challenge CarMax and try to gain more market share. They need to execute this ADESA play perfectly. If they do, Bill Nash and his team should be worried.

Indeed. No one can deny that Carvana has done well in the past few years, even though they were close to a disastrous situation. What's interesting is that they're now posting the best gross profit per unit numbers in the industry, even as their volume is decreasing. The big debate is whether they can maintain this level of productivity and profitability per unit while growing. Or how much will they have to give back? It seems like CarMax and Carvana, especially CarMax, have a clear advantage. There's little risk around whether CarMax will succeed. They can be more methodical and profitable over time. Carvana, on the other hand, is a high variance situation. They were hypergrowth, building infrastructure and accruing fixed costs to grow. They've spent about two years shedding as much third-party involvement as possible, using their first-party infrastructure built for much larger volume growth that didn't materialize. They're now at 20% or 30% utilization across all their asset classes, whether it's IRC, parking lots, short haulers, long haulers, etc. They're likely to start growing again within the next six months. So, how do they turn on growth again?

How do you manage that? Clearly, there's a staffing requirement for these IRCs and every operational link in the chain. More inventory needs to be acquired, reconditioned, and transported, which requires significant staffing. Advertising is also necessary to attract customers who want to sell their car to Carvana or trade-in. However, I have a broader question. If they have a $2,000 margin to work with compared to other market players, why don't they adjust the pricing on both ends to win outright? Do customers not care that much about price? If it's a difference of $30 a month to their payment, how significant is that to them? I don't have the answers to these questions. I'm unsure how much they can leverage these economic advantages to gain real market share.

That's why I always say this about them. I believe they're at an inflection point. It reminds me of what happened with Best Buy. Best Buy had to make a tough decision to eliminate all their commission associates and switch to hourly wages, which hurt them because customer service declined. I think Carvana has done the same thing. The only difference is Carvana cut and shed all this infrastructure, all this additional overhead and expense that they deemed unnecessary. They've let go of many people over the last couple of years. I question how they're going to scale now. They've only brought in a fraction of the people from ADESA, not all of them. They've also let go of many of those. Now, they're trying to expand further. How do they do that while maintaining customer engagement, satisfaction, and securing repeat sales? That's challenging because it requires more than just the desire to do it. It demands a strategic mindset. One thing Carvana still lacks, which CarMax does exceptionally well, is post-sale service. That's a crucial part of the customer lifecycle, whether it's a CarMax customer or a traditional franchise dealer customer. Many of us prefer to take our cars back to where we bought them if there's an issue. It's human nature.

If I were Carvana, I'd focus on three things. First, how do I build out this scale at ADESA? What infrastructure do I genuinely need? Not technology, but the actual infrastructure required to sell these cars. What human capital do I need? Is it people for reconditioning, moving the cars, or selling the cars through my online setup, my e-commerce platform? Lastly, I'd be considering who I could partner with for post-sale service, other than my sister company, DriveTime.

No consumer associates DriveTime with a great customer service experience. It's a lower-tier car dealer that caters to subprime paper. No one wants to go there. If I'm a prime paper, I don't want to sit in that waiting room with my $60,000 or $80,000 Lexus while they're servicing their $10,000 Ford Focus.

It may sound harsh, but it's the truth.

Indeed, you're absolutely correct.

They don't seem to have a partner for this. I don't believe they need to build their own service team as it would be too costly due to the need for service centers. However, they do need a partnership. Perhaps with a company like Pep Boys, although it might not be the best choice. They need a program that reassures the customer.

Consider a CarMax customer's journey. They select a car, secure financing, and sign the documents. Before they take delivery of their car, they are introduced to the service team. They meet the service manager and the service rider. They know who to call for an oil change. This entire experience is part of the process.

In contrast, with Carvana, you insert a token, your car comes down, and you drive off. There's no introduction to a service team. What happens if the car won't start two days later? Who do you call? With CarMax, you already know where to take it, even if they delivered it to your driveway.

That's why I believe Carvana has the potential to seriously challenge CarMax. They already have been, but they could take a significant portion of CarMax's current market share, not just the share CarMax hasn't taken yet. They could continue to take CarMax customers, but they need to clearly communicate what the Carvana buying experience entails.

If you gather a group of 100 CarMax buyers, they will all tell you a very similar story. However, if you gather 100 Carvana buyers, you'll hear a multitude of different stories. The experiences vary greatly, whether the car was delivered, picked up from the car tower, or if there was a service issue.

I question how much they've really cut costs, as their numbers look significantly better compared to twelve months ago. It's difficult to make such drastic changes in this industry. I wonder what they've actually done, as sometimes the information they provide can be embellished.

Yes, it's often embellished. I believe they were overpaying for cars, offering 100% to 200% of the car's value to ensure they had the inventory they wanted. They probably offered free delivery and transportation when they shouldn't have, to increase conversion rates. By not doing this anymore, they could save $500 to $1,000. I also think the standardization of the checklist in the recon centers wasn't as thorough as it is today.

I believe you're correct.

It seems they underwent a significant transformation in creating decision trees and checklists.

They were urged to get their act together.

When I took my tour two months ago, it was evident that they were implementing new decision trees and formulas. It seemed like they were overwhelmed with growth and were doing whatever they could to keep up. They didn't run a tight ship from the start. Perhaps they did initially, but chaos ensued when they tried to grow at a rate of 100% per year.

This growth strategy is completely foreign to the CarMax management team. When you try to grow at such a rapid pace, things start to break. However, the broader question is whether, with enough standardization and systematization, and with the infrastructure to handle a surge in volume, you can provide a consistent experience.

Can you ensure there's not much variance in the quality of the units, the economics of the units, and the customer experience? Can you grow your share and brand in the market and get the customer flywheel going, like CarMax has done? They could greatly benefit from a service portion of the business, probably through a partnership, to care for the customer after the sale and keep them engaged with the brand throughout the lifecycle of their car. I know we've gone over time, so I apologize.

The market is challenging because it's dominated by CarMax and Carvana. They control the market, and most people agree. However, when they see the actual market share these companies have, they're often surprised it's not larger.

The private party car market is difficult to penetrate, and it's become even tougher with Covid. Despite the pandemic, many people have been able to sell their cars on online marketplaces.

I have a conceptual question. Companies like CarMax and Carvana have excellent checklists for inspecting and reconditioning cars. Why can't they enter the peer-to-peer market? For instance, if Aaron wants to buy a car from Marty, they could pay a fee to have the car reconditioned to CarMax's standards. CarMax could facilitate the transaction and even arrange financing, without having to own the inventory. They could perform all the work for a fee, arrange profitable financing, and let the peer-to-peer transaction happen. Is this a possibility?

In theory, yes, it could happen. What you're describing is essentially the real estate model that we're familiar with. However, the reason it hasn't been implemented is due to the sheer scale and cost of building it out. The initial investment is so substantial that no one is willing to take the risk, as you won't see a return until you earn it back.

Let me give you an example. CarMax partnered with Kelley Blue Book to create the KBB Instant Cash Offer program. This was designed to drive buyers into the CarMax showroom. You could go online, get your KBB Instant Cash Offer, print it, and even send it to the nearest CarMax. However, it didn't work because we couldn't get our teams behind it.

In most car sale environments, you're dealing with a commissioned salesperson. There's a sense of pride and ownership in the sale that prevents a shift to a different model. Until everyone is willing to move away from this, and adopt a model similar to Best Buy's, it's unlikely to happen. No one is willing to risk the capital, especially when no one else has done it.

Another factor is the current state of the real estate market. More and more agencies are leaving the Realtor Association, saying they can do it on their own. This is similar to why CarMax isn't in NADA or NIADA, and why Carvana hasn't joined them. They don't need these organizations to sell.

I have some friends at a large group in Texas, Sewell, who have been focusing on creating an environment where they don't have to pay commission. They tested this in a few of their dealerships years ago, but it failed. The reason is that salespeople are used to getting different amounts of money for different cars. As long as this mindset exists, it won't change.

It would take a complete shift in approach, like what CarMax is doing with their Customer Experience Centers. If CarMax were to take it a step further and do what Carvana is doing, blending the best of both models, then we could get there. However, this would mean eliminating the need for salespeople and physical locations.

If we could combine the smooth, seamless transaction of Carvana's model with everything CarMax has done, we could get close to what you're suggesting. But the question is, is anyone willing to take that risk? It would be a game changer and would require a group of pioneers to implement it. It might even be better to bring in people from outside the car industry, as they might be able to see beyond the traditional way of doing things.

I love the idea, but the execution is the challenge. The risk of capital is so great that no one's willing to put it out.

This has been an enjoyable conversation and I've learned a lot.