1157: How to Improve Processes, Remove Friction, and Accelerate Innovation with Jon McNeill

Pete Mockaitis
Jon, welcome!

Jon McNeill
Thanks. Nice to be here, Pete.

Pete Mockaitis
Well, I’m excited to discuss The Algorithm and I’d love it if you could kick us off with a story you shared, which I found riveting, in which when you were just getting started at Tesla, even actually before you had the job, you took it on yourself to tackle a project. Could you tell us the story?

Jon McNeill
Yeah, I was talking with Elon about joining Tesla and I kept saying to him, “I think you need a big company guy to come in.” And he said, “No, that’s exactly what I don’t need. I need a fellow entrepreneur.” And I said, “Well, I want to make sure I can be helpful to you. So, like, what’s your biggest problem right now?”

And he said, “We have a demand problem.” And I said, “Okay, like frame that for me. They’re public companies. So what’s your commitment to the street this quarter?” He said, “Roughly 12,000 cars.” And I said, “How are you doing so far?”

It was about a month into the quarter. He says, “It’s a month into the quarter. We’ve sold like 2,500 cars, short of 3,000 cars.” I’m like, “Oh, now I understand your demand problem. You’re going to miss your quarter. Let me go to work on that and see if I can help.”

So I went to, just in the travels with my own business at the time, I went to eight different cities over the course of about 10 days. And I went into eight different Tesla stores and did what I was told was the pinnacle of the sales process and sales funnel, which was taking a test drive.

And the whole idea was that people would come in, they would get in an electric car for the first time, hit the accelerator. And the accelerator in an electric car is way different than a gas car. It’s instant torque to the wheels, and so you take off like a rocket. And so a lot of people experience that. And then they can’t stop thinking about it until they’ve got a car.

So I went in eight stores, I did a test drive. And then, oddly, like a few days later, I hadn’t heard from any of them. So Elon had put me in touch with his head of sales ops. And I called him and said, “Like, am I blacklisted in the system or something? I used eight different email addresses so people wouldn’t know who I was. And people don’t know who I am anyway. And I’m not getting any callbacks. Why?”

And he looked in the system, he’s like, “No, you’re not flagged or anything.” I said, “Can you do me a favor? Can you tell me how many test drives you’ve given in the last 30 days that haven’t been followed up?” And he’s like, “Sure, give me, like, an hour. I can go run that.”

So he calls me back in an hour, and he said, “Nine thousand, 9,000 test drives, no callbacks.” I said, “Well, congratulations. You’re going to miss your quarter because you haven’t called anybody back. And no wonder you’re so far short of orders.”

So then I said, “Can you shut off a store rep’s ability to take any new leads until they’ve called all of their previous test drives back?” He said, “Yeah, I could do that.” I said, “How fast?” He said, “Globally? I could do that in a few hours.” I said, “Awesome. Do it, because we got to force-change, like, super fast or, otherwise, you’re missing your quarter.”

So he does it. Calls me back the next day, he’s like, “You won’t believe the orders are flowing in like crazy.” And I said, “Yeah, because you’re just calling people back. It’s the easiest thing ever. You’re asking for the order.” And then it dawned on me that I didn’t work for Tesla yet. And so I kind of gulped, and I said, “Hey, I got to call you back, because I got to call Elon and beg for forgiveness.”

So I called Elon and said, “Hey, look. Here’s the situation. You’re demand-challenged. Here’s what I found out. You’ve done 9,000 test drives with no callback. And so here’s what I did about it. I was on the phone with your head of sales ops and we shut down any new leads into the system until everybody called their previous test drives back.”

And I said, “But I got to ask your forgiveness. I was acting like this was my company because I’m a CEO in my day job. But this is not the company I’m the CEO of. You are. This is your company. And I did something and I didn’t even ask your permission. And I’m really sorry. I got to apologize.”

And I didn’t know it because we were just getting to know each other. But I had this long period of silence on the other end of the phone line, and I thought, “Oh, my God, what have I done?” And it seemed like forever. It was probably 60 or 90 seconds of just dead silence.

What I’ve learned since is that’s how Elon processes. He, like, shuts off all other sensory input and just thinks. And so I was about to ask him if he was still there, and he hopped on and he said, “You know what? That’s exactly a rational decision, and I’m so glad you made it. I think you’re going to fit in here just fine.”

And that was the last hurdle. I said, “Well, I think I’ve proven myself I can be useful. So I think I’m ready to sign on, too.

Pete Mockaitis
That’s a lovely story on many dimensions. But when I reflect on it, what I’m stuck with is, “All right, Tesla, great, great brand, tremendous company, tremendous leaders, lot of smarts there.” How is it that we find ourselves in a situation where one of the most fundamental things one can do in a sales operation – call people back – isn’t happening?

Jon McNeill
I think I see this over and over again. People look at a lot of data and miss the obvious. And when you just go and see a process with your eyes – this is going to sound completely old-fashioned – but it’s actually the case.

Like, I believe that the strongest analytical instrument you have as a leader are your eyes and ears. And if you go to the front line and eat your own dog food and experience the product or the process for yourself, you’ll often see exactly what’s wrong. But unless you know the exact piece of data to look for that would give you the clue, you don’t know that.

And so this is a complete hack I’ve found in leadership is just go to the front line, go experience the process yourself. You’ll understand really rapidly exactly what’s wrong. And it doesn’t always work, but it works a lot of the time. And this was an example of that.

So a bunch of really smart people looking at data missed it because they didn’t leave their offices. And my encouragement to people, and those people, as I then became their leader was, “You’re going to leave your office at least one day a week, and you’re going to go to the front lines because you’re going to see stuff way faster than the data will show you. And, hopefully, we’ll miss a lot of these potential big divots in our plan if you do.”

Pete Mockaitis
Yeah, that’s tremendous. We had Marina Nitze on the show, who was the CTO for the United States VA, and that was one of her core recommendations, is follow the thing all the way through the process and then you’ll say, “Oh, that’s why it’s taking forever because so-and-so is waiting for something to be faxed. Well, let’s stop that. All right, now we know. Understood.”

All right. Well, so then zooming out a bit, your book, The Algorithm, any particularly surprising or fascinating discoveries you’ve made in your career or while putting together the book that you’re putting forward here?

Jon McNeill
I think this is, like, I had gone into Tesla having been a student of the Toyota Production System and Lean, and what I found was those frameworks were awesome for incremental improvement and optimization, but not awesome for quantum growth.

And so what we tried to distill in The Algorithm, this was the whole team at Tesla and, really, the leadership team that came up with this framework, was to distill a framework for quantum growth into some digestible steps that we could push to the edge of the organization.

And the edge could start to innovate super fast, because that’s what doubling a business every eight months at scale requires. And that’s exactly what the team was pulling off. They were doubling a business every eight months.

Pete Mockaitis
All right. So the classic Toyota manufacturing, Lean, Kaizen, Kanbans, etc., is great for your small incremental growth and you’ve got a whole another thing for hyper growth. And can you unpack sort of basically what is the algorithm?

Jon McNeill
It’s a five-step process that we used on a weekly basis to drive innovation. And this is used at Tesla. It’s used at SpaceX. And I think, as leadership experts or academics go to study Elon 20 or 30 years from now, and ask the question, like, “What made this guy such a successful entrepreneur and industrialist? Like, what was it?” I think they’ll come down to this, like, weekly cadence of deploying this framework, because that’s exactly what he does.

He determines the one or two issues that are existential to the business. In other words, the two things that could kill you if they don’t come true. And then he just devotes his time to that. He delegates everything else to the team. And so you have great agencies as a leader because you’re running the business and he’s really working on two existential issues, but he’s doing it on a weekly basis.

And what that does is that keeps the organization innovating on a weekly basis. So if you’re doing that every week, you’ve got 52 opportunities to build advantage versus your competition. And that’s exactly what that delivers.

Pete Mockaitis
Okay. Well, so could you maybe walk us through a couple examples of, all right, grab existential issue and then work through the steps?

Jon McNeill
So I’ll take you through one to make this super simple. So the first step is question every requirement. Second step is you delete every step and you simplify. The third is you optimize. The fourth is you speed it up because no process can get faster and have high quality with speed unless it is like optimized, so that’s a test. And then you automate last, which sounds kind of nuts especially coming from a technical company or Silicon Valley company to say automate last, but I’ll kind of illustrate that.

So let me start with the first, the negative example. So we started to prepare for Model 3 production. And the idea of the teams was, “We’re going to have a fully automated line, the most automated manufacturing line ever in the history of automobiles.”

And so you might remember this phrase, “We’re going to build the alien dreadnought,” or, “the machine that makes the machine.” That was this whole era. And so teams went to design the Model 3 production line in digital simulation. Today they would call it a digital twin. And they designed all the machines, the flow, the conveyors, the people, everything. They didn’t lay it out manually and do it manually first.

And so when we were down to the wire in terms of, “We got to produce Model 3s because we’re going to be out of cash if we don’t,” they went to start that line, and the line wouldn’t work for a thousand reasons. There were major, major mistakes that were made in that digital SIM.

And digital SIMs usually don’t work because they can’t think of everything. And cars were falling off the conveyors, falling off the line. And that then led to a radical step, which was one of the leaders, Jerome Guillen, said, “I’m actually going to scrap this whole process and build a tent outside. And we’re going to do what we should have done from the start. We’re going to build cars by hand.”

And so we started, over the weekend, building Model 3s by hand. And we simplified the entire process because we didn’t have conveyors, we didn’t have a lot of the machines that were necessary, and we did the bare essentials to produce that Model 3.

So we simplified the process, we deleted all the unnecessary steps, we didn’t have the luxury, really, of having many steps. And then we started to optimize the process, and we started, we produced 50 cars a week, and then we produced 100, and then we produced 500. And the goal was to get to 5,000. And we kept creeping up, creeping up, creeping up by optimizing the process and applying speed to it.

Once we finally had the process nailed, then we automated at the very last step and moved from the tent back into the factory. We had rebuilt the factory production line by this time so that it could actually produce cars.

And when we went to do the postmortem, we said, “How would we have avoided this? Number one, we automated first, not last, and it almost killed us. We did not run the process manually first. We did not delete a bunch of steps. We did not optimize the process. We automated before we had done any of that work, and we almost killed the company.”

So it was, really, at the end of Model 3 production, that the algorithm came together, and we said, “Here are the steps we’re going to follow from here on out. When we go to launch a new product, when you go to invent a new product, we’re going to follow these steps.”

And so that’s an end-to-end example of how not to do it, but then it led us to a framework of how to do it.

Pete Mockaitis
Okay. So maybe digging into each of these in some detail, with “question every requirement.” I have a feeling many of us, it doesn’t even occur to us to question the requirements because the requirements, they’re just like the air we breathe or the water we swim in. It’s like, “Well, this is just how it is.”

So could you maybe surface for us some prompts, or some examples, or some perturbations so that we can identify some requirements that maybe need to get questioned?

Jon McNeill
So when we started to sell cars online, we figured out it took 64 clicks to sell a car, but 44 of those clicks were in the financing process. And that’s because a loan doc or a lease doc for a car has 12 pages of paragraph after paragraph after paragraph. You’ve got an initial sign, all this stuff.

And so a question on requirement step that we did was I took that loan doc to our general counsel at the time, his name’s Todd Maron. I said, “Todd, out of these 12 pages, I need to get rid of some clicks. Can you tell me how many of these paragraphs are a requirement of law or regulation?” He came back about 24 hours later and he said, “None.”

I said, “Well, how does a 12-page document exist then?” He said, “It’s the result of a bunch of well-meaning corporate lawyers at banks trying to protect their client. And so they insert paragraphs for all these uncertainties or these edge cases that could come up to protect the bank.”

But he said, “None of these are required by law or regulation. And, actually, the bank has case law on its side. Like, they don’t need all this stuff because if somebody cheats in a certain way, that’s already been decided by case law. So it’s not like you need to make somebody acknowledge something that the courts have already acknowledged as being out of bounds.”

So I said, “Todd, you’re telling me we could have a one-paragraph loan or lease that says, ‘Here’s the price of the car, here’s the interest rate, here’s the term, and here’s the monthly payment?’” He’s like, “I’m telling you, you can do that.”

Nobody in the industry had questioned whether or not the loan doc or the lease doc should exist. We were just crazy, silly enough to make that, to raise that question. We discovered something that none of our competitors had discovered, which is you didn’t need to put customers through a 44-click process. And as you know, if you had to buy anything on Amazon, and it took 44 clicks, you’d probably opt out a lot. And that’s true in e-comm, and it’s certainly true when people are buying a $100,000-car online.

So we went to talk to banks to see if we could get anybody to go along with us on this one-click loan release, and we got the door slammed in our face by everybody, even though they would intellectually acknowledge that they didn’t need all these paragraphs.

They said, “I would never take the risk of doing this. I would never take the career risk of going to my CEO and suggesting we do this.” Finally, we found a very enterprising CEO at US Bank, and he said, “We’ll do it. In fact, we’ll talk to a CrossTown digital bank called Ally. They’ll probably do it too. We’ll take your loans, they’ll probably take your leases, and we’re off to the races.”

All that started because we questioned why a 12-page loan doc had to exist. And that’s probably the clearest example or the best prompt I can give you on questioning requirements. There’s a lot of stuff that nobody ever questions and takes as a given. If you question that stuff and it turns out it’s not a given, now you’ve got an advantage.

Pete Mockaitis
Yeah, I like that a lot. And I’m thinking about, well, the phrase that comes to mind is checkbox marketing, or really checkbox anything. If your mindset is, “Well, that’s just what’s done. It’s just what needs to happen.” And I’m thinking about, well, I have a one-page guest release form.

And I think lawyers tend to want to put a lot of intense, you know, things in there. It’s like, “I own all of these forever, no matter what.” And I think, “Well, actually that’s not the vibe I’m trying to put out there when I’m just meeting somebody. I’m just sort of like, ‘Hey, you and I are both cool to do what we want to do with this thing, okay? So no pressure, it’s all good.’” And that’s kind of what I’m trying to convey.

So it was not the default. We had to shift and adjust and then it is, well, it’s much smoother because then I don’t have a lot of people say, “Whoa, whoa, hold on, buddy. What, what is this? I don’t know. Wait. Time out. I got to talk to my lawyer. I got to talk to my agent. I don’t know if we can do this, you know?”

No, that just about never happens because it’s very quick and it’s simple and it’s handy. And I liked that example because the law, in particular, feels like something that’s immovable, like, “Oh, ‘legal’ said, we just have to have that. And end of discussion. It was like, “Oh, well, maybe legal would be willing to have a follow-up conversation and see what can be done here.”

Jon McNeill
Totally. And it takes a certain mindset. So, like, Todd, as a leader of the legal teams was willing to come along on that journey with me and question requirements, and that’s pretty rare in that kind of a leader in that function, but he was super commercial and business-oriented.

And so he would start that journey without having a bunch of hesitation. Like, he’d say, “Yeah, let’s go, like, look into this and see if this is really true and if we really need this.” I don’t experience many general counsels like Todd because they largely go into that career to mitigate risk. And so that’s where their position starts.

Pete Mockaitis
Well, and as a point of curiosity, I mean, I understand frictions and clicks are, you know, undesirable. They slow down the behavior you want, and this is cool and convenient and nice and pleasant to have fewer clicks.

But, you know, Jon, really, lay it on me, is a few dozen clicks for signing an initialing a hundred thousand dollar vehicle enough to move the needle so that folks are like, “You know what? I wanted that Tesla, but this is too much effort. I’m done with this document. I’m abandoning my cart”?

Jon McNeill
It happened a lot, and when we eliminated it, the opposite happened. People started to buy like crazy online, and our digital sales went through the roof, which is because it’s just science. In e-comm, clicks equal anti-conversion. So you get rid of clicks and your conversion rate goes up. It’s just math.

So this is why when you get to the cart in Amazon, the search bar disappears. They want to take away any potential click that you’re going to do other than hit order, because they know how hard it was to get to that point and how hard it is to get you to actually click the order button.

Pete Mockaitis
Yeah, and I believe they’ve litigated the one-click ordering.

Jon McNeill
Exactly.

Pete Mockaitis
It’s like, “No, this is ours, we own it, no one touch it.” And it’s like, “Very touchy,” and because the stakes are huge.

Jon McNeill
Because of that, because it’s powerful. Yeah.

Pete Mockaitis
All right. Well, can you chat with us about deleting the steps in the process?

Jon McNeill
Yeah, so here’s another example, and Elon actually cut a little video in this early on, but he had taken on this challenge in manufacturing where there was this separator between the battery and the car. And so it’s basically along the floor of the car.

And we could not produce this separator, big piece of plastic. We couldn’t produce it to save our lives. It was warping and it wouldn’t fit right, etc. And so this is one of the chapters where he’s literally sleeping in the factory trying to solve this problem.

And, eventually, he asked the team, “I got to talk to the engineer who spec’d this part. Who spec’d the part?” And the closest people around are the battery people, and they said, “Oh, it’s the auto dynamics team. They wanted a noise dampener in between the battery and the passenger compartment.”

So he grabs the guy that’s the head of the auto dynamics team, and said, “Why did you spec this part?” He said, “I didn’t spec the part. It was the battery team. The battery team told us they needed a heat shield between the battery and the passenger compartment.”

He’s like, “I was just with the battery team and they said it was you, not them.” He said, “Give me the name of the person who spec’d this part.” So they go look for the name. And it turns out the name of the person that spec’d the part was a summer intern that didn’t even work at Tesla anymore.

Elon had spent weeks in the factory trying to solve this problem, all for a part that didn’t need to exist. And so, at that point in time, “We said, nobody claims ownership of this ‘requirement.’ So we’re going to delete it out of the car. And, therefore, we’re solving a whole production problem that was holding us up.”

And you will find, over and over again, that there are steps you can delete from your sales process, from your delivery process. And the hack to finding those is, essentially, map your process on a wall with a bunch of sticky notes. Then have your team go circle those steps that the customer pays you for. It turns out there’s very few of those.

They don’t pay you for the order sheet. They don’t pay you for the PO. They don’t pay you for, in our case, the bill of lading. They don’t pay you for all this stuff. And the things they don’t pay you for are immediate candidates for deletion, because you’re doing those for internal reasons, and you’re creating cost for internal reasons.

Now some of the stuff is necessary to track dollars and cents. I totally get it. They don’t pay you for accounting. They don’t pay you for tax. I get that. But there are a bunch of steps you can cut out when you start to say, “Hey, the customer really doesn’t pay us for this. And we don’t really get anything out of it. So why are we doing this?” And those are good candidates to delete.

Pete Mockaitis
That’s good. Well, I was curious, did Elon get on the phone with the intern? Like, “Hey.”

Jon McNeill
The intern didn’t even work there anymore. Like, didn’t even work there.

Pete Mockaitis
Well, mean, hunt him down wherever he is.

Jon McNeill
He was like, “I don’t need to waste time talking to the intern, like, it’s gone.”

Pete Mockaitis
Okay. So then simplify and optimize.

Jon McNeill
Yeah, so once you delete steps, now you’ve got way fewer steps. So you’ve got a simplified process and now you can start to optimize that process and make it go faster and faster and faster and faster. It’s a little bit like starting any skill, whether that is playing a musical instrument or a sport, where you start and you’re like, “Man, I can’t go very fast.”

And then you practice and practice and practice, and it turns out you get faster and faster and faster because you get more and more efficient and optimized. And that’s the idea here is you keep speeding the process up and speeding it up.

Now you’re still in manual mode. So you’re learning a bunch about what’s causing, what’s getting in the way of speed, and the stuff getting in the way of speed is usually a quality issue. And so you eliminate these quality issues. And then once you’ve got the target speed achieved, now you know you’re optimized and you can then start on automation.

Pete Mockaitis
A quality issue in terms of something needs rework or is outside of the spec we’re looking for?

Jon McNeill
Exactly.

Pete Mockaitis
All right. And then with the acceleration, so there’s practice. What are the other drivers of accelerating?

Jon McNeill
Mainly, it’s putting speed goals that are double the current speed. So you say, “I want to either double the output at current speed or I want to double the speed at current output,” either one of those works in my book, just try to double.

And it starts to reveal, “Oh, here’s all the things that are in the way of doubling. So maybe we can delete more steps. Maybe there are more requirements that we should question. Or, maybe there’s a better method.”

And as you do those, then you start to identify, “Ah, there are a bunch of things that we didn’t see at first that we now see now, that we can either delete or simplify to help us speed this process up.”

Pete Mockaitis
So the doubling of the speed is not so much a, “Hey, it turns out we can just double the speed, it’s all good.” But rather, it’s sort of like a magnifying glass to identify, “Oh, that’s what’s holding down the speed doubling.”

Jon McNeill
Totally. There’s this great series of scenes in the Hulu series, “The Bear,” which is about this super high-end restaurant in Chicago. And it’s turned from a roast beef cafe with the same team in the back that is now trying to earn its Michelin star.

And there’s a woman that’s at the pasta station, and she has two minutes once an order comes in to cook pasta. And she starts at five minutes. And then she does things like she portions each serving of pasta in a little plastic container. So she’s pre-portioned.

And she starts to do multiple boiling water pots to drop the pasta into so she can do more than one at a time. And then she’s got a saucier step that she realizes, if she preheats the sauce, it meets the pasta at the right time, she can speed it up. And she gets the process down from five to two and a half minutes by just eliminating all these steps, but she’s still not at her two minute target.

And then the sous chef, who really runs the kitchen, comes over and is just helping her. They have a busy night. And so the sous chef comes over and does a few things that experienced chefs know to do. Bam! Bam! Bam! Order comes out in a minute and a half.

And the woman who is at the pasta station says, “I’ve been working this problem for like 90 days and I got it from five minutes to two and a half. And you walk over and you do it in a minute and a half.” And she’s like, “Yeah, I’ve done this before. So just watch me and you can see the extra steps you can take out of your process to make yourself faster.”

That’s what we’re talking about is just optimize, optimize, optimize until you get to a really different output than you’ve had before.

Pete Mockaitis
And then the final step is automation.

Jon McNeill
Yeah, and then you can automate it. Because automation is like wet cement. When you put it around a process, it sets up really hard and you got to get jackhammers to get rid of it. So you don’t want to automate until you’ve really got the process nailed. Otherwise, you’ll be suffering with the cement that’s already solidified the current process where it is.

Pete Mockaitis
Okay. So now, as we were talking about this with the restaurant example and manufacturing examples, I’d love to change gears a little bit and hear about, let’s say there’s someone who’s in the middle of an organization and their output is not physical, but, nonetheless, there is a process that’s cranking out something on the other end. Could you give us an example in that domain of working through those steps?

Jon McNeill
Yeah, so, like, whether you’re producing software or whether you’re producing marketing or you’re producing financials, I think the same steps apply. If you step back and say like, “Hey, look, I produce financial statements for the business,” or, “I produce a budget for the business.”

If you step back and say, “Okay, let me just map the process that we use today to do that. And maybe I should question some base assumptions here. And do we have to do all these steps? Like, are all these things really required?”

And so an example of that is Tesla, when I joined, was doing standard annual budgets. And the senior management team and the CEO would set the target for sales, and the target for gross margin, the target for cash flow. And so these annual budgets were being done in the way they’re done at almost every company.

The challenge was we were doubling the business every eight months. And so, like, if you’re trying to project out a budget for a year, it was horribly inaccurate by, like, month three, because things were changing so fast. So all that was essentially wasted effort, all that planning and budgeting.

So we asked ourselves a question, like, “Time is super, super valuable in this company. How could we improve this process? And rather than going through a big planning cycle every year, what actually needs to get done?”

And where we ended up was, we questioned the requirement of having a year budget. It turned out what we did was we had quarterly budgets because the business was moving on a quarter basis, not a year basis, and we would have these rolling four quarter budgets.

We would spend less than two days setting targets in the budget because it was only going be good for a quarter. So you didn’t want to spend a week of the quarter, a huge chunk of the quarter doing it. And that finance team evolved a whole different way to do financial planning and budgeting.

And they did it, not on an annual basis, but a quarterly basis. They did it on a rolling basis, and they just used the key inputs that everybody was looking at in the business anyway, “What’s the sales rate? What’s the production rate? What’s the margin?”

And we can build a budget off of that, and rather than taking the whole company’s time planning, we’re just going to keep rolling this and rolling this, and we’re absorbing everything that we’re learning from the market as we double.

And so you can apply this whether you’re sitting in a finance department or a marketing department or a sales department. You can apply these principles to your advantage, too.

Pete Mockaitis
Okay. Well, Jon, tell me, any final top tips, do’s or don’ts, for folks looking to implement this stuff?

Jon McNeill
As we talked about, like one of the principles that is a complete hack is eat your own dog food. I learned this from Sam Walton in his book, Made in America, where he would tour stores on a weekly basis to find out what the customer wanted that they didn’t have, what they had too much of that the customer didn’t want, and what were the things that store managers were doing that were super good that he could spread across the company.

And I learned from that and started to see that, “Oh, he’s onto a complete hack.” Like, if you use your own product, if you experience the customer experience, you’re going to see all kinds of areas of improvement. And if you teach your people to see it before you, they’re going to move even faster than you can.

And so I would say, like, the secret hack is go experience your own product and use it. We even went so far at Tesla to have a rule that you couldn’t present product in a PowerPoint. We didn’t want to see, like, some rendering or rendition. We wanted to see the real thing. So if you’re presenting product, you had to do a live screencast onto the screen so we could see the product, we could play with the product, and we could see how it actually worked.

I was with a group of bank executives a few months ago. I asked them to raise their hand if they actually use their own bank’s consumer app. No hands went up. And I said, “I could have guessed that because I’m a consumer of two of your banks, and your app sucks.”

“It’s so bad that if you used it, you wouldn’t live with yourself for another day without fixing it. You would call up the head of engineering and you say, ‘We got to fix this, this, and this.’ But you don’t use it.” So the organization gets the sense that nobody cares. And if you can live with it, they can live with it.

And eating your own dog food changes all that. It creates an immediate feedback loop to the top and it allows you to set the bar of acceptability with your organization.

Pete Mockaitis
Okay. Thank you. Well, now could you share a favorite quote, something you find inspiring?

Jon McNeill
I think this whole approach is about simplicity, and so I love the Mark Twain quote, “I would have written you a shorter letter if I would have taken the time.” It speaks to how hard it is to simplify. Humans, I think, we are naturally complicators. We’re not natural simplifiers, and it actually takes work to simplify.

And I love that quote, because it reminds me, each time I read it, of the fact that simplification is work. It’s super valuable and rewarding work, but it’s work.

Pete Mockaitis
All right. And can you point to a favorite study or experimental or bit of research?

Jon McNeill
I think, if you want to understand AI and the current version of AI that we have with LLMs, the best piece to read is the original DeepMind paper.

Pete Mockaitis
“Attention Is All You Need”?

Jon McNeill
Yup.

Pete Mockaitis
All right. And a favorite book?

Jon McNeill
Favorite book would be, I’ve got two right now, one is The Goal, which taught me a completely different way of looking at business, by Eliyahu Goldratt. And the second is Unreasonable Hospitality by Will Guidara. I think it applies to every business. It’s so good.

Pete Mockaitis
Okay. And a favorite tool?

Jon McNeill
Reading. Like, if I could answer the question that way, I’d start every day reading for an hour and a half, and I read variety of things. I read books, I read, obviously, the news. I read Twitter, I read Reddit, Hacker News sometimes. Reading for me is a tool.

Pete Mockaitis
Okay. And is there a key nugget you share with teammates, audiences, that they quote back to you often, a Jon original?

Jon McNeill
We had the standard for service at Tesla that gets quoted back to me, it got quoted back to me today actually, and that is, “Make them talk about you at dinner tonight.” Do something that is so awesome that they’re going to talk about you at dinner tonight.

Pete Mockaitis
Okay. And if folks want to learn more or get in touch, where would you point them?

Jon McNeill
I’d point them, you can find me at DVX.ventures.

Pete Mockaitis
All right. And do you have a final challenge or call to action for folks looking to be awesome at their jobs?

Jon McNeill
Become a simplifier and you’ll stand out.

Pete Mockaitis
All right. Okay, Jon, thank you.

Jon McNeill
You bet. Thanks, Pete.