Austin Ligon co-founded CarMax in 1991 as an executive at Circuit City and served as president and CEO of the spun-off company until 2006. Today he is an angel investor and advisor to retail, service, and educational startups.
Q: Where did the idea for CarMax come from?
We were looking for areas that no big-box retailer had put effort into. Automobiles were obvious, but new cars are heavily controlled by franchise laws and manufacturers’ rules. Looking at used cars, there was almost no data beyond the big picture that more than half of the value and the majority of cars sold were used cars. It was an industry that no one thought of as an industry. It was a business most people look down on, even considered sleazy. We saw a free-market opportunity to address a huge consumer problem.
Q: Why the fixed-price model?
Quakers invented the fixed-price store. Their idea was, if I tell you the price for something is eight, but I really only want six, then I’m lying and that’s a sin. I have Quaker ancestry through my mother. That sensibility resonated with me.
Before CarMax, I was screwed by every dealer I dealt with in every way you can imagine. Like everybody else, what I wanted was the car. I didn’t want all the rigamarole—the hide-the-number pricing, financing gimmicks, upselling, or haggling. I’d give them my keys to get my old car evaluated for a trade-in, and I couldn’t get away because they wouldn’t give them back. All of those things that drive us crazy, they’re intentional. Dealers know how to take advantage of it and use it to their benefit.
Bringing a big-box retail experience to buying a used car was a concept anyone might have come up with. Why wouldn’t you sell cars that way? The challenge was executing on the idea effectively enough to tap into scale and network effects.
Q: What are the economies of scale?
There are economies of scale from having a brand name and training salespeople with a common sales process the way that Home Depot, IKEA, or Costco would. There are economies of scale in promoting the brand and shopping experience through advertising and marketing. Those were new to selling cars because there weren’t large used dealers. And the new car business was designed to be fragmented.
Humans tend to find the problems that are easy to fix and profitable for them, and they tend to not find the problems that are hard to fix and not profitable.
That meant we had to figure out the front end. How do you create a fixed-price concept that consumers like? How do you manage pricing? How do you manage an honest sales force? Those things are complex, but other retail businesses have had to figure them out too, so we were adapting rather than creating those systems.
The back end—the acquisition and reconditioning of huge numbers of used products—that was unique. Operationally, I’d say it’s by far the most complicated business in retail.
Q: Would you explain the back end?
The back end of used cars means you’re essentially in the acquisition and remanufacturing business. We had to create a system for doing that. It was a learn-by-doing, iterative process that took us at least a decade to get to the point that we really knew what we were doing and could do it efficiently in a large-scale, replicable way.
We had to develop a buying team that knew what used cars are worth. What should you pay for maybe 800 different models of cars in all kinds of conditions? How do you avoid the lemons? How do you deal with warranty and guarantee so that when a lemon does make it through the process, you take care of that? And once you’ve bought the cars, how do you get them where you want them? The transportation logistics of moving lots of cars around and doing that at the lowest cost is difficult, but we eventually turned it into a strategic advantage. What nearly put Carvana under twice was figuring out how to buy cars effectively and then how to transport them effectively.
The biggest challenge was reconditioning cars. We never knew which cars we’d end up buying and reconditioning in any given week, so we couldn’t pre-order parts. The economies of scale with reconditioning only emerge when you turn it into a manufacturing-style process, but there were endless unforeseen problems that kept us from getting to that steady flow we wanted.
Q: How did you approach it?
It was our biggest headache area for years. When we bought a car, whether it was at auction or from a consumer, we paid out the cash that day. Early on, it took us two to three weeks to recondition a car. We were supporting a huge amount of inventory that was taking up space and earning us nothing while waiting to be reconditioned. We knew we needed to dramatically increase the speed of the process while dramatically lowering the cost and improving the quality.
For a long time, one technician would diagnose what was needed and then make the fixes. We kept tweaking that process. It got better and better, but we weren’t getting the dramatic improvement we needed. Eventually we hired an industrial process engineer from FedEx. He helped us understand when that one-technician system was as efficient as it could ever be. And that the system had a key flaw—a lack of objectivity in diagnosing the problems.
It turns out that humans are humans. They tend to find the problems that are easy to fix and profitable for them, and they tend to not find the problems that are hard to fix and not profitable.
Throughout the industry, technicians are paid on a piecework basis. Some fixes can be done a lot faster than the standards say, and some are a lot slower. The technicians all know which is which. Brakes are faster and thus more profitable. Alternators are harder and time-consuming. We were getting more brakes replaced than needed and not enough alternators fixed.
Conceptually, from day one, we knew the right way to do this was to have a diagnostician whose only job is to find the problems, and then another group of people to fix those problems. We just couldn’t figure out how to do it.
The real breakthrough happened when we started working with the Toyota Production System Support Center. They teach Toyota’s process and quality improvement systems to other companies, but you can’t just hire them; they have to accept you as a student. It took us about two years to persuade the head of the quality group to visit. We were told to expect him to spend 30 minutes at one of our locations. He spent all day. His team said, “That’s good.” And as he was finally leaving, he said, “Interesting problem.” His team said, “That’s really, really good.”
We had convinced him that not only were we sincere about wanting to do a great job of reconditioning late-model used cars but also that the problem was more complex than he initially believed and therefore interesting.
We separated diagnostics and fixing. We changed how we paid technicians. We built a predictive system for ordering parts. We developed a manufacturing-like process where one person works on brakes, one works on alternators, etc., and it flows down a line like you would in manufacturing cars.
It took 10 years to get to the point that we were making real progress and then another 10 years to get to the point that huge benefits flowed out of that. But we went from two to three weeks to recondition a car to a few days and were fixing the problems that need to be fixed and not fixing things that didn’t need fixing. That made the business a lot more profitable. We’re still iterating. The Toyota folks say, “We’ve been at this 50 years, and we’re still not even halfway there.”
Q: CarMax buys cars at auction—but it also sells a lot of cars at auction. Why?
In the retail business, CarMax sells late-model used cars, but we buy any car a consumer has the legal right to sell us. Which means we buy lots of cars that are older than what we want to sell. We had to figure out how to get rid of them. Doing that efficiently was another critical behind-the-scenes part of making CarMax work.
For many sellers, when they send cars to a wholesale auction, if bids don’t meet their minimum reserve price, they won’t sell. If GM brings a thousand cars to auction, they may only sell 70%. Our view is that these are bananas, not fine wine: they’re not getting better with age. If we have a hundred cars, we’re going to sell a hundred cars. Whatever the price is this week, that’s what we sell it at.
And where most industry auctions require sellers to disclose 2 or 3 things about the cars, we tell buyers 15 different things. We tell them exactly what’s wrong with every car. We want everyone to know what they’re buying.
Our customers are mom-and-pop car dealers, because we offer a steady supply of cars and everybody is clear on what’s for sale. It’s local in each market, so people don’t have to travel. That all gets us the fastest turnover and the highest bids.
It turns out, if you run a very efficient auction, you make a few hundred dollars on every car you sell, with no investment in reconditioning and very fast turnaround. By a return-on-invested-capital basis, it has become a very profitable business.
“The Yale School of Management is the graduate business school of Yale University, a private research university in New Haven, Connecticut.”
Please visit the firm link to site