The Secret to Out-Innovating the Competition: Inside the Tesla Playbook

Motley Fool analyst Rachel Warren talks with former Tesla President Jon McNeill about the five-step formula for achieving hypergrowth, the hidden metric every investor should track, and the AI revolution.

To catch full episodes of all The Motley Fool's free podcasts, check out our podcast center. When you're ready to invest, check out this top 10 list of stocks to buy.

A full transcript is below.

This podcast was recorded on April 5, 2026.

Mac Greer: “I wanted people to see, like, what was behind the scenes driving this success of Tesla, who seems to out innovate their competitors again and again and again. And I wanted people to see that that could be done by frontline people, everyday people. You didn't need to be Elon Musk to do this.” That was Jon McNeill, former president of Tesla and author of the new book, The algorithm, the hypergrowth formula that transformed Tesla, Lululemon, General Motors, and SpaceX. I'm Motley Fool producer Mac Greer. Motley Fool analyst Rachel Warren recently talked with McNeill about that formula and about why you don't have to be Elon Musk to out-innovate the competition. Enjoy.

Rachel Warren: Hello, everyone, and welcome back to Motley Fool Conversations. I'm Motley Fool analyst Rachel Warren. Today, I'm excited to welcome Jon McNeill to the show. Jon currently serves as the CEO and co-founder of DVX Ventures. Previously, Jon served as president at Tesla. Following his time at Tesla, he joined Lyft as COO, where he played a pivotal role in doubling the company's revenues, helping to take the company public. Jon currently serves on the board of numerous companies, including General Motors, Lululemon, and Stash, and his upcoming book, The algorithm, the hypergrowth formula that transformed Tesla, Lululemon, General Motors, and SpaceX. Jon shares behind the scenes look at how iconic companies scale and outlines a playbook for leaders to drive sustained growth and impact. Jon, welcome to the show.

Jon McNeill: Thanks. Nice to be here.

Rachel Warren: Excited to talk with you today. One of the things that's really interesting, I want to start off talking about your book. You've codified this formula that has been applied at a range of these companies that I noted called the algorithm. I'd love it if you could walk us through your book and its themes, but really also talk about those steps of the algorithm in order and what it means for companies.

Jon McNeill: The algorithm is really the operating system, I would describe it, that got invented at Tesla really through making mistakes. We made a lot of mistakes, then we would do postmortems and say, how the heck did we get here? Then we would develop a principle to not do that in the future. That's the heart of the beginning of the framework. The framework then was used, once we developed it, to give everybody in the company a framework from which to innovate, really on a weekly basis.

The reason I wrote the book was I wanted people to see what was behind the scenes, driving this success of Tesla, who seems to out innovate their competitors again and again and again. I wanted people to see that that could be done by frontline people, everyday people. You didn't need to be Elon Musk to do this, which is why the framework was developed because Elon can't be everywhere all the time. We had literally thousands of people that were driving innovation all over the world who were following this framework. The framework to the heart of your question is basically five steps, and then I identify three secret ingredients that really make it work. But the five steps are question everything at the beginning. Question the requirements that you've been given because you want to not build a business around a bad set of assumptions, or you don't want to build a process around a bad set of assumptions. I'll give you an example of that from the financial market. My firm today we invent companies. We start companies ourselves and grow them and scale them. We took a look at the ETF market and looked it like supercycle ETFs and found that the top holdings of almost every ETF in the supercycle space was Amazon, Google, Microsoft, Nvidia, and Apple. You can buy that as an investor for two beeps. That's called the Mag 7 or the S&P 500. You can buy that product. You don't need a supercycle product at 75 BPs doing the same thing. But we step back and started to question the requirements of why that was, why ETFs were built this way.

The first thing you get a requirement that everybody who builds an ETF follows, and that is they build ETFs by market cap weightings. If you want to do an AI infrastructure ETF, and you're going to follow the market weighting assumption that the whole industry follows, you drop Nvidia into that ETF, and it overwhelms every other stock because its market cap is so high. But if you take a different approach, so we started to say, is it a requirement to formulate ETFs based on market cap? The answer is no. It's not a requirement of law. It's not a requirement of regulation. It's just a requirement that has been evolved over time because that's the way everybody does it. We said, that doesn't give investors exposure to profit pools. There's a lot that goes into an AI data center that's not Nvidia. It turns out there are 60 public stocks that go into an AI data center. If you weight those public stocks by their contribution to the AI Data Center, you get a very different-looking ETF that actually exposes investors to the actual profit pools. This is a unique situation where we've got five years of commitments by the hyperscalers to AI Data Center builds. We know what the purchase orders look like. We know it's going into those so that we can backwards engineer that and say, which stocks are going to benefit?

It turns out when you create an ETF, and you question the requirement for market cap weightings, and you say, is there a better way? You come out at the other end with an answer that says, yeah, there is a better way. That is in this case, to weight by profit contribution. We developed an AI infrastructure ETF based on profit contribution, not market cap. The industry told us we were crazy. Nobody does it this way. We said fine, because we had back tested it to know that we beat the stuffing out of every other AI infrastructure ETF on the market. We introduced that in December 2025, so just over a year ago, and it is over the past 12 months, the top performing AI infrastructure ETF in the market. You would have been up more than 80% if you'd bought it on day one. Because it turns out that it reflects the profit pools that are getting created. As the market wakes up to these individual stocks that are being helped by AI Data Center builds, the value gets created in those stocks and our ETF takes advantage of that. But that's an example, the very first step of the algorithm, which is question everything. Because when you start to question things, you start to discover there are these assumptions that people operate under that are really self-limiting. If you step out of those self-limiting assumptions, you can create an innovation that creates a lot of value and breaks new ground. That's the first step of the algorithm is question everything, basically.

ADVERTISMENT: In January of 1915, Ernest Shackleton's ship Endurance became encased in the ice in the Weddell Sea. Through determination, grit, and savvy, Shackleton would lead his men through a brutal winter, then over hundreds of miles of Antarctic ice, followed by 800 miles across some of the roughest waters in the world. It is one of the most extraordinary and inspirational journeys in the history of exploration. Find this story and many others at the Explorers Podcast available wherever you get your podcast or at explorerspodcast.com.

Rachel Warren: I was going to say, can you walk us through the remaining four.

Jon McNeill: Then the next one is a second step in simplification, and simplification is hard. Most people don't do it. There's that famous Mark Twain quote that I would have written you a shorter letter if I'd just taken the time. It's because simplifying is hard. It takes a lot of work. The first step of removing false assumptions and requirements is a step to simplify. The second step in the algorithm is the second step of simplification, and that is delete every step you can in a process, and you probably haven't deleted enough until you've had to add some back. Delete ruthlessly. Then the third step of the algorithm is once you've got this simplified process, now you manually run that process. This has got to sound crazy coming from somebody who spent their entire career in tech and hard tech, not to automate. But we tell our teams go manual because all the best businesses were built manually first.

The original team at Amazon didn't automate the distribution centers we see today. They actually went and bought books. They put up an order site. They would take an order. They would go buy the book from a local bookseller, put it in a box, ship it so that they learn the distribution side of the business before they automated anything. The founders of DoorDash, who were CS majors at Stanford, put up a PDF of restaurant menus with a phone number at the bottom, and the phone number rang in their dorm room, and then they would order the food from the restaurant, go pick it up and deliver it so that they could simplify the process. We tell our teams do what Amazon did, do what DoorDash did, do it manually first, because when you do, you learn firsthand all the opportunities to simplify. That's the third step of the algorithm. Run the process manually.

Fourth step is now speed it up. Put cycle time constraints around it because when you speed something up, it exposes all the quality flaws and process flaws. Now you're further refining. Then once you've got this process optimized with speed, the fifth step is automate last, which again, sounds totally counterintuitive coming from a tech firm, a place like Tesla, but we just learned this lesson so many times where we would automate first. When you automate a bad process, all you do is speed up the time to a bad answer or a bad outcome. We learned over time to automate last. Probably the sharpest example that happened in Model 3 when we were trying to design a factory that was the most automated factory in the world. You might remember Elon talking about the alien Dreadnought and the machine that would make the machine. There was an attempt to design a factory completely digitally, before it was done manually, before the process was actually run in the real world. When that happened, when the machines were installed, you could see that, oh, my God, huge mistakes were made. Like, there wasn't enough room in between the machines for humans to get in and maintain them. You couldn't even maintain the machines. That automated line never worked. That was close to $1 billion mistake. When we did the postmortem on that and said, How did we get ourselves into this situation where we had to build a tent in the parking lot, to build cars manually, to actually save the company with cash flow that was desperately needed, how did that happen? It was because we automated first, not last. We learned these principles by making the mistakes along the way. That was a doozy of one that led to that last point.

Rachel Warren: During your tenure at Tesla, you saw the company go 2-20 billion in 30 months in the early years of the Model 3 launch. I wonder, was there a specific algorithm break through that moved the Model three from those production doldrums to a cash flow machine? Was that driven by superior product, a superior operating system, or both to apply your framework there?

Jon McNeill: Kind of both. We had to start to apply the algorithm to each problem we were facing, whether that was how to do service in a business that's doubling every eight months with an installed base that's doubling every eight months. Anything from service to how the car was built to how the car was sourced to how the car was designed. Like, we took this framework and gave it to teams. An example of that was we were selling cars faster than we ever had in our history, and we couldn't build service centers fast enough to service the cars. We got a small team together that ran our service business. Actually, the team that ran our Palo Alto Service Center because it happened to be just down the road from headquarters, so we could work with them really quickly and iteratively.

We went to the manager of that site and said, "Can you figure out how many cars you can fix before the companies or the customers done with their cup of coffee?" He's like, "Yeah, why?" I said, because a customer can finish their cup of coffee, say, in 20 minutes. If you can fix a car in 20 minutes, we may not need a building for that. If we don't need a building, we may be able to do service in a way that has never been done in the industry before, and that's mobily in customers' driveways or their offices. The guy said, "Yeah, Jon's accepted, I'm up for it." Worked on that problem for like a month. He came back a month later and said, you guys have to come see this. We went down the hill. What we saw was a parking lot that was buzzing. The parking lot had three lanes that cars would go down. A customer was met at the entrance to the parking lot by our senior tech, like the senior surgeon in a hospital. That senior tech would ask them for the symptoms of what was wrong with their car. Because that senior tech had seen everything, they knew whether that car was going to be a small, medium, or large repair. Those were the three lines that the car went down in the parking lot. The smallest and mediums could be fixed in the parking lot, and the largest had to go inside. It turns out that 80% of the cars didn't have to go inside. We didn't need a building. If we didn't need a building, we could fix those cars anywhere, not just in a parking lot at the service center. We could fix them in the customer's driveway or home office or in their parking lot of their office, where they went to work.

Questioning the requirement first office, do we have to do repairs in a building? The entire industry believes that. Dealerships have enormous service centers that they've built. Midas and Meineke and others have built enormous footprints of buildings to fix cars. We question the requirement of do you need a building to fix a car? The answer was no. We ran the process manually in a parking lot to get it working. Then we took a couple of hundred Model Xs that had been returned on lemon law returns, stripped out the insides, put tooling on the insides, put technicians in those. It turned out we answered the question, could a technician be more productive if they didn't come to a building or at least as productive? The answer was yeah, we found out that they could. We started to deploy these just in San Francisco and the Bay Area, and customers were blown away that we would come and fix their car and their driveway, like magic elves. We tried to make it fun. We put espresso machines in these Model Xs so you could get an espresso while your car was being fixed. Just had a bunch of fun with it.

Then the last step was we automated all the process around scheduling and parts of workering and all this stuff. It fundamentally changed the way that automotive service is done. Still, car manufacturers cannot figure out how to do this. They could go try to copy, but I think until they ran the process manually within their own system, they probably wouldn't optimize it for themselves. But that's an example of how a service team, just by using this framework, completely innovated in service, and that didn't involve Elon Musk at all. That was just really diligent people on the front lines following this framework to invent something new.

Rachel Warren: Something you pointed out and something I also took away from your book was this idea that really the speed of innovation is such a key factor. I think it begs the question, does a company with a shorter cycle time tend to have a more durable mot than one with just a strong brand? How does this apply to areas outside of tech?

Jon McNeill: I think one of the things that Japanese taught us, and Toyota taught us was this metric that Toyota executives follow, which is cash velocity through a business. We all learn to evaluate investments based on market cap metrics like EBITDA multiples, revenue multiples, which have growth metrics built into them. But what the Japanese do is they say, really, the measure of how good you are as a leader is how fast cash moves through your business. An example that is, when I started at Tesla, it was roughly 14 days for us to take a pile of aluminum on one end of the factory and turn it into a car on the other end of the factory. Toyota at that time, could do that in about four days. They could take a pile of aluminum, and Lexus would come out the other end of the factory or a Toyota or a Scion in four days. Took us 14 days, took us four days. What does that mean? That means Toyota can do the same amount of business with a third less working capital than we could. Let that sink in. Speed of cycle time and speed of cash mean that you need far less working capital than your competitors. If you need far less working capital, you have a balance sheet advantage. You have a liquidity advantage versus your competition. That's why cycle time of cash or speed of cash really, really matters. It's really hard to measure that, and so that hasn't become a measure that is available to the street or to investors. But it's absolutely critical. If you want to assess how good you are versus your competition, it's almost like the 40-year dash for businesses. Like, what's your time?

Rachel Warren: Can a culture of speed be retrofitted into a slower growth company or is it something that's more of a DNA level trait?

Jon McNeill: I think it's a DNA-level trait. There's this old adage. It's hard to make a sprinter out of a marathon because the sports are so different. That's because the muscles get toned up and trained to go 26.2 miles versus 100 yards or 100 meters. I think it's just hard to take a business that hasn't been measured on cycle time or has been a slower moving business and speed it up. Not impossible, but just you got to know what you're getting into if you're biting off that challenge.

Rachel Warren: We've talked a lot about how the algorithm applies to tech sectors, but I want to lean a little bit more into its application to other sectors outside of tech. I mentioned earlier on the boards of GM, Lululemon. How does one apply the algorithm to say, a century-old automaker versus a high-growth retail powerhouse? How can we as retail investors view those elements when we're evaluating investments?

Jon McNeill: I think the first clear evidence of this is top-line growth. You look for top line growth. But then you've got to achieve that top-line growth with a discipline and that discipline is reflected in gross margin. That's the first discipline indicator that you get is, are they doing this really well? Then third, you get operating expense leverage over time because you're not adding headcount for every new dollar revenue that comes in, and so you're getting operating expense leverage. That comes down to EBIT and operating cash flow. Taking INVIDIA for a second. They got the biggest market cap because they're generating high two-digits, three-digit growth, so incredible growth at that scale. At a close to an 80% gross margin. They're showing operating discipline and the ability to continue to command price out of their customer efficiency out of their organization, and then they're also getting operating leverage, so they're kind of three for three. As an investor, I'd look at that and say, I have industry-leading top-line growth. I have an industry-leading gross margin, and I have industry-leading operating cash flow. That looks like three reasons to say yes to that stock, depending again on how far it's run past its peers and how much upside there is, but that's the first indicator that, yeah, this is something you probably should consider investing in.

Rachel Warren.: Just a couple more questions for you. One, looking ahead to the next decade, what are some industries or sectors, even that you think are best positioned for an algorithm style disruption?

Jon McNeill: I think, wealth management. It befuddles me today that consumers still have to put together their own estate plans and tax plans, and those are disconnected from their investing plans and often lead to adverse outcomes because they're disconnected. There are not 1,000 ways to construct a portfolio for a person a certain age in a certain stage of life. I do think the wealth management industry, which has largely been a sales-driven approach of going and getting clients and assets is going to change, or you could say it in a more negative way, it's going to get disrupted because there's an ability to deliver more sophisticated outcomes that are coordinated for customers and much higher value for customers. That's one where we have our eye on it, and we're going to try to maybe play in that space and figure out how to do that, too, because there's $1 trillion transfer that's happening between generations over the next 10 years, and that's going to create a bunch of wealth management opportunities in this current manual bespoke process. Is it going to serve that kind of capital well?

Rachel Warren: Is there a sector or industry that looking ahead to the next 5-10 years excites you the most?

Jon McNeill: That's one that does in terms of the opportunity for it. I think AI is one of the first technical revolutions that's happening in affecting white collar work first, not blue collar work. Every other technical revolution has essentially affected blue-collar work first. I think we look across white-collar sectors for opportunities because this is a unique moment in time where there's going to be disruption and value creation happening at scale in white collar. There are a bunch of industries that excite us as we look with that kind of filter on.

Rachel Warren: One last question. If someone listening or watching this could only take away one or two algorithm metrics to really track the health of their long-term holdings, what should those be?

Jon McNeill: Health of their long-term holdings. In terms of metrics in those companies, I would look at cash velocities, so cycle time, and the ability to expand margins over time, both operating margins and bottom-line margins. That tells you whether management teams are really good at what they do.

Rachel Warren: Fantastic. Jon, it's been so great to talk with you today. I wish we had more time. Those were listening or watching. Check out Jon's book, The Algorithm: The Hyper Growth Formula, The Transformed Test, Lululemon, General Motors, and SpaceX. It's a fantastic read. Jon, thanks so much for joining me today.

Jon McNeill: Thanks, Rachel.

Mac Greer: As always, people on the program may have interest in the stocks they talk about, and The Motley Fool may have formal recommendations for or against, so don't buy or sell stocks based solely on what you hear. All personal finance content follows Motley Fool editorial standards and is not approved by advertisers. Advertisements are sponsored content and provided for informational purposes only. To see our full advertising disclosure, please check out our show notes. For The Motley Fool Money team, I'm Mac Greer. Thanks for listening, and we will see you tomorrow.

Mac Greer has positions in Alphabet, Amazon, Apple, Lululemon Athletica Inc., Microsoft, and Nvidia. Rachel Warren has positions in Alphabet, Amazon, and Apple. The Motley Fool has positions in and recommends Alphabet, Amazon, Apple, Lululemon Athletica Inc., Lyft, Microsoft, Nvidia, and Tesla and is short shares of Apple. The Motley Fool recommends General Motors. The Motley Fool has a disclosure policy.