How Nvidia Built a $1.7 Trillion Company by Breaking EVERY Rule | Ep 257
20:04 Share

In 1993, Nvidia had zero revenue. Today, they are at a $1.7 trillion market. This was not luck. This is a strategy that no other company has done, a strategy that you can apply to scale your business. And so in this video, what I want to do is I want to share with you the five unconventional methods that allowed Jensen Huang to turn Nvidia from niche chip maker into the most important technology company that exists on this planet.

Each method I'm going to break down is going to build on the one before it. So let's start breaking them down. Most CEOs and founders operate with a layer of secrecy. They meet with their key executives behind the closed doors, and they control everything that their employees do or don't hear. Jensen hates that. And the reason for that is because

At most companies, decision-making is this black box. And at Nvidia, it's an open book.

In fact, he even gives one-on-one feedback in front of the entire team. I can speak firsthand from my experience that this is a better way to do things. Now, to the degree that they show transparency, I think that's based on your preference as a CEO or a founder and your business that you have, as well as the size of it. But what I can say is this, I remember that when I first had my company, Gym Launch, I had experienced leadership come in and they were selling me. You don't wanna share the P&L. You don't wanna share the revenue. You don't wanna talk about the profits.

You don't want people to know all these things. I remember there was one day where I was like, I really feel like this is holding people back because they don't understand what numbers they influence in the business. They don't understand what the scoreboard is for if they're winning or they're losing. And so finally I said, you know what? I'm going to teach everybody on my team how to read a P&L. And so I pulled all of my managers in and I taught them how to read the P&L. And that single meeting was one of the most important meetings I could have ever done because everyone from there on out felt a sense of ownership over the results of the business.

And it wasn't that all of a sudden people were like, why don't I have more money? I should be getting paid more. Look at the profit. Look at this. That wasn't what happened at all. It was actually just that they took a greater sense of ownership over those things. Now I can say from another point of view, to go from one-on-one communication to one-to-many communication is one of the most advantageous things that you can do as a founder.

A lot of the time what happens when you start the company, it's built off of one-on-one communication because it's small. That's what works. And it's the easiest way to communicate. So what happens is as the company gets bigger, you keep communicating like that. And what I think that he's figured out here is that at some point you get so big that

it is not beneficial for you or anybody you're talking to to have so much one-on-one communication. Because the same questions and areas where one person lacks clarity, especially in a company of his size, there's probably 50 to 100 other people who lack clarity or have questions around that thing. And so what it does is it says, great, every time someone has a question, every time there's an issue, I'm just going to project it to the whole company. Because not only are you making sure that other people don't have those questions,

but you're also letting the whole company know where you stand on different issues. So they start to understand more of how you think. I understand there are guardrails in place. And so some of these things, you don't necessarily have the liberty to decide. But for example, if you're a small bootstrapped business and you're like, how could I apply this to myself?

ask yourself and look at your calendar, how many one-on-ones are you having a week? In those one-on-ones, are you teaching those people things that you could be teaching your whole team? That's the first place I would start because it might be that instead of having 15 one-on-ones, you might need to have a group meeting and three one-on-ones. Whereas if you're a big company,

I think that this actually even applies to a greater degree, which is look at, if you're the CEO, all the skip level meetings you're having and ask yourself, do I really need to have all these skip level meetings or do I need to have training? That's just like me explaining to people clarity of this area, talking about my philosophy here, talking about my philosophy here. A great question to ask yourself is if somebody in your company makes a mistake, does the whole team learn?

Or does just that person learn? And you can replace the word mistake with does something well, asks a question. It's just that if one person on your team learns something, how do you make sure you get to apply that learning to everybody else? If your team spends more time managing relationships to get information than doing the actual work, you probably have a transparency problem.

Transparency is a game changer for sure, but it's still not enough. Most businesses still get stuck. NVIDIA doesn't. Now, why is that? Because they killed the biggest bottleneck in business. And it goes against everything that you and I have been taught in leadership, which brings me to my next point. Flat teams move faster. Most CEOs have five to 10 direct reports.

Jensen Huang has over 50. The more direct reports a CEO has, the less layers are in the company. It allows us to keep information fluid. It allows us to make sure that everyone is empowered by information.

And our company just performs better because everybody's aligned. Everybody's informed of what's going on. - This was mind boggling to me when I saw it because I'm not gonna lie, it gave me a ton of anxiety. I was like, how would I inherit all the direct reports of my company? But then I said to myself, you know what? I'm gonna take pieces of what I like from what I'm listening. And so here's what I learned. At Nvidia, there are no managers of managers.

His direct reports aren't just executives. They include engineers, researchers, operators, and they're all working on NVIDIA's core products. Now, the reason that Jensen can operate NVIDIA this way is because he understands level four delegation. So let me walk you through the four stages of delegation as I like to define them.

them. The first level is investigation. That's when you're gathering research. So essentially, somebody else goes to get the information for you, and then they bring it back to you for you to make the decision. So an example might be, hey, go research three manufacturers for our product line and bring me a report. That would be level one. Now, level two is informed progress. They execute your looped in on updates.

So an example for that might be, I want you to set up our influencer marketing campaign and I want you to check in with me at each of these milestones so that I can give you feedback. And then level three is informed results. That's where they complete the project and then they tell you the outcome. So an example of that would be something like, launch the new product line. I don't need updates. Just bring me the results and the P&L when it's done. Level four is complete ownership. So you hand over full responsibility to somebody, no check-ins needed. Example of that is just like,

this is your department now, run it as if it's your own company. The reason why I believe that this works for him is because everybody in his company is somebody who can handle being delegated something at a level four. If you look at the people in your company, for example, where this would go wrong is that you hire people who don't have experience, who need training, who haven't run their own product line before, and then you say, go do this. It's not like when you have 50 direct reports, you have time to train them on their job. He's bringing in probably the best of the best in the world.

So what I think you can take away from this is understanding that if you bring in the best of the best in the world, you want to give them more autonomy. But if you're at the point in your business where you can't bring in those people yet, you're still going to have to train them. You're still going to have to work with them. You're still going to give them more time. Now, the second piece of that that I've thought a lot about is that in some companies, there's two sources of reinforcement that you can get.

You either get reinforcement from the boss or you get reinforcement from the work itself. At NVIDIA, they're working on something that's literally changing the world. People want to work there just to be able to say they helped build that product. They don't need a pat on the back for the work that they're doing because they get a pat on the back from seeing it's changing the world. They get a pat on the back from the fact that they get to put

this on their resume and say that they helped build something that changed technology. If your product doesn't do that, it's going to be very hard for you to get away with managing the same way, which is essentially not doing much managing at all. Because again, they need more reinforcement from the manager because they're not getting it from the product. So here's what I want you to think through. Do your managers make it easier or harder for you to run the business?

Do you have more or less transparency because of the managers? You should have more transparency when you have managers in place because they should be feeding you more information more frequently and it should make it easier for you to understand what's going on in the department. That's the first thing to think with. Now, the second thing to think with is if you have people who are very autonomous, you brought them and they have a lot of experience, are you level four or at least level three delegating to them? Or are you treating them like somebody who's at a level one or level two? What I want to challenge you to do

is identify if there are people who need to be at a level four and push them up at least one level, give them more ownership this week. Especially if you're a smaller company or earlier in building your business is that you keep people at a level one or level two and you need to be moving them up the ladder. Think about it like this.

How many layers exist between me and the real problems in my business? If there's too many layers between you and the problems, if you don't know what the problems are and it's really hard to figure them out, there's too many layers. If your leadership team needs constant managing, you don't have a leadership team,

you have a liability team. And so you need to reassess if those are the right people to be leaders in your company. It doesn't mean they don't have a seat in your company. It just means that they might not be the best leaders. What I will say is this, the flatter the team, the faster you typically move.

But speed without truth is chaos. If the real problems don't reach leadership, then decisions get made in a bubble. And that's also why Nvidia uses what is called the top five system. So really interesting, being so close to the action in his company, Jensen actually does not do status reports. - The reason why I don't is because status reports are, you know, they're meta information by the time you get it, you know? And so they're barely informative. - Yeah. - You know?

And so I've distilled and refined and bias has been inserted and perspective has already been added. And you're not looking at ground truth anymore. Instead, every employee at NVIDIA at any level can email Jensen directly with their five most important observations, insights, or concerns. I really love this philosophy that he has. Does that mean I don't have status reports? I do have status reports.

Maybe I am not evolved enough in my CEO career. However, what I will say is this, is that I think what's so important about what he does here is he essentially has an open door policy with his email. He's saying you can reach out to me directly. And I actually encourage it because I want you to get your information from the source. That is something that's really important in building a company because especially as you're growing quickly, sometimes your managers don't have the best information or the most up-to-date information.

Sometimes they're just trying to learn on the fly as well. And sometimes they're busy doing other stuff, or maybe you don't feel comfortable asking the question. And so something that I've tried to employ in all my companies, like you can always Slack me, you can always reach out to me. You do not need to feel like you need to funnel something through your manager to ask me a question. If I think your manager will answer it better, I will tell you that and I will direct you to them, but I will not not answer your question. I do think that like having an open door policy as a CEO is really important if you want to grow your company into being something excellent.

I think a lot of people worry, oh my gosh, is this going to step on managers toes? Is this going to? The question at the end of the day is this, is this going to make it more or less likely that you achieve your goals as a company? And I think that it makes it more likely when people feel like they can speak to the CEO who's determining the vision of the company. I think that's going to make it more likely that you're going to hit your goals. I think a great question to ask yourself is, do my employees know what the company's priorities are?

How often am I seeing gaps in what they think the priorities are versus what I know the priorities are? And if you're seeing gaps there, I think it's a great idea to make yourself more accessible. I said this to one of my employees the other day. He brought a problem to me and I solved it in like five minutes. And I said, gosh, why don't you bring more problems? I can solve them in 20 minutes tops. And he texted me this morning. He was like, you know, I'm still thinking about that. You said that because I've been dealing with this problem for weeks. Your strategy isn't what's on the slide. It's the talent that you trust to execute it.

In fact, your talent is the strategy. And so in order to make sure that they're aligned, they need to hear it from you. Now, getting real time insight from your team is incredibly powerful when it comes to growing a business. But what you do with that information is going to be what sets you apart in the industry from everybody else.

At NVIDIA, Jensen doesn't just listen. They adapt and they adapt really, really fast. They don't waste time on long-term plans that are just gonna be set in motion to make everyone feel good and then not adhere to. Instead, they use what they call continuous planning. Now, a lot of companies of the size of NVIDIA, probably not anybody watching this YouTube video,

And if you are, hello and welcome. The companies of that size, for the most part, they spend thousands and thousands of hours on consultants to draw up like five to 10 year plans. At NVIDIA, they don't actually even have a single long-term plan. - We don't do a periodic planning system. And the reason for that is because the world is a living, breathing thing. And so we just plan continuously. There's no five-year plan. There's no one-year plan. There's no plan.

There's just what we're doing. - This is how that changed their trajectory with the AI breakthrough in 2012. So researchers at the University of Toronto built an AI model, it's called AlexNet, and that dominated the global image net. That was powered by Nvidia GPUs. So what did Jensen do? Instead of treating AI as a niche use case, he redirected Nvidia's entire engineering force

towards AI computing, personally meeting with researchers and ensuring that their GPUs became the industry standard. So what was the result of that? Well, we all know what this company is today. By the time like Google and Microsoft and Intel realized that AI was the future, they had already built the entire ecosystem, cementing themselves as the backbone of AI computing. So it's interesting because people, especially in interviews, candidates ask me all the time, they're like, what's the vision for acquisition.com? Where do you see it in 10 years? And I'm like, listen, I can give you a guess, but

But the reality is, is that none of my plans for 10 years out have ever been exactly how I planned them to be. Now I can have a vision of like what I would love to accomplish and some key milestones, but do I have a plan for 10 years? Fuck no. The farthest out that I will plan is three years. And that is even a stretch. I try to focus mostly on the one year because I say, okay, I'm going to take the three, I'm going to engineer it back into one year, but

What's most important to me is the one year and not even that, but the next quarter. Because the reality is this, we love to have this illusion of control, that we can control our business and control all how it's gonna succeed or fail and that planning is gonna allow us to do that. And so I will say one thing, which is like, I try to speak a year out and I try to plan

based on what's happening in real time in the world. For example, this last quarter, I said, you know what, fuck it. I didn't have a huge investment in technology I was making in the company, but looking at the landscape of everything that's happening, I need to build out an entire engineering and product department for acquisition.com. Now, did I have that in my plans? Did I slot that in my hiring? No, but I was like, I wanna take advantage of what's going on right now. I wanna make sure that we're technology first. I wanna get us on the cutting edge of what's happening.

And so because of that, fuck my plans. I'm going to do what's going to work now. The first thing I would say in terms of what we can take from this is ditch rigid five 10-year plans. Anyone who I know who hits their five or 10-year plans is probably a really fucking lame plan. It's probably really small thinking and it's probably something really easy to hit no matter what the fuck happens.

Sorry if that's the case, but I just don't think that for big shit, if you want to do things that change an industry, if you want to change the world, if you want to build something amazing, shit's going to change. And you just have to be able to make real-time decisions based on market and economic shifts, as well as the shifts in what your customers want, because that's changing in real time. They don't give a fuck about your five or 10-year plan.

They give a fuck about what they see in the market, what they want right now, how they're feeling, if their wallet is full, if it's empty, if it's light, if it's heavy. And so because of that, what I think the second thing we can take from it is to shorten our feedback loops. One thing that one of my first mentors said that I will always stick with me was like, we don't wait until the end of the quarter to reassess our priorities. We should be reassessing our priorities on a weekly and biweekly basis. I am constantly looking at the priorities of the company and thinking, are these still the top priorities?

And I'm not doing it every month and I'm not doing it every quarter. I'm doing it every week. And then what I can do based on that is that I'm not trying to hit a goal that doesn't matter. I'm trying to make sure that I can make micro pivots based on new information or new opportunities in a fast way. And so what I would say is like, don't wait for a quarterly or annual planning session to change your business. Do it in real time.

And a great question you can ask yourself is, am I following a plan that no longer fits the needs of my customer? Am I following a plan that no longer keeps me winning in the environment that it is now? Adapting fast keeps you ahead, but speed alone is not what builds greatness in a company. So then the question is what actually does? That is what brings me to the last and probably the most important point in this whole thing.

thing, which is that high standard start at the top. Employees all over have described Jensen Wang as demanding, as a perfectionist, as not easy to work with. His response?

It should be like that. If you want to do extraordinary things, it shouldn't be easy. He's not asking his team to do anything that he wouldn't do. He talks about all the time. He's there from 5:00 AM to 9:00 PM every day, including weekends. And the results speak for themselves. His team is incredibly bought in and Nvidia has some of the highest retention rates in the entire industry. Not everybody likes working at Nvidia, you know, and people have heard me say, say that Nvidia is an institution, but we're not a church or a prison. Not everybody

gets to come not everybody has to stay but the people that do love working there because we give them terrific projects we give them extraordinary resources and we also give them teammates that they know they can count on if you're very demanding of people and your company only exists to generate money for you to put in your pocket good luck they're all going to quit

For real. If you're very demanding of people and you treat them very well, and you have a mission that goes beyond making money, and you're pouring all the money back into your company, I think you have more room to be demanding. And then to a greater degree, if you exist to make the world better, if you exist to save lives, then you better be demanding because you're not going to achieve your mission unless you are. So I truly have thought about this a lot, and I really believe that the amount of demanding that you can be of people is related back to why your company exists.

Of course, we're demanding of heart surgeons because if somebody dies because you aren't demanding of them, it's on you. But if the shareholder makes less money because you weren't demanding, right? And so it's all about what's the mission, what's the impact of your company, and how does that translate back to the people on the front line? And something I've learned from myself is that I allow myself to be more demanding when it's not about me. My first company, a lot of the reason it did exist in the very beginning was like, I didn't really know anything. I was just like, oh, to make money?

and to help people at the same time. In the very beginning, I didn't really understand. And then once I felt personally secure and had money and paid off debt and all those things, then it was like it changed. And then I felt the ability to be more demanding. With acquisition.com, it's not about the money at all because I became financially free with my other company. I took all the money from that and put it into this one. And so if anything, I'm risking more of my money here. And so of course I'm more demanding of people because I'm saying I want to build something bigger than me, bigger than this building, bigger than everybody here. I want to build something that goes beyond us.

And I don't know exactly what it looks like yet. I think it's got this three year vision. I think we're going to go here. But I feel like I have the ability to be more demanding when it's not for me, it's not for them. It's for a bigger reason. Now, I think what we can take away from this, I think, is really important. The first one that's irrefutable is caught, not taught. You have to lead by example. You have to set the standard yourself.

There's a reason that I come in to my office every day at 6:00 AM and that I leave only before I go home to eat dinner and go to bed early because I'm a grandma and because I want to be back here at 6:00 AM. And the reason is because I don't want to ask people to work harder than I work. I want them to know that I'm working. I want them to see me working. I want them to see I'm in my office. Your work ethic, your attitude, the way that you show up creates the culture, whether you want it to or not. So if you want to raise the bar, get out there and raise the fucking bar. You do it with your actions, not with your words.

So I think a question that you could ask yourself is, am I leading in a way that creates a team that grows or creates a team that's happy?

Growth and happiness do not come from the same things. And the thing is, is a leader's job isn't to make people comfortable. It's to make them better. At its core, NVIDIA follows three principles. Move fast by keeping information flowing so everybody from interns to executives, they get access to the same truth at the same time. The second is stay lean and efficient. There's not wasted time on like status reports or approval chains or internal politics.

They're trying to cut through that bullshit constantly. And then the last one is attract the best talent because the smartest people

people want to work on problems that haven't been solved before that go beyond the shareholders the employees or even the company themselves and that is why Nvidia is out maneuvering trillion dollar competitors it's because they saw the AI Revolution coming years before the rest of the world and why the best of the best always want to work there because the companies that win are not just the smartest companies they're the ones who can execute faster than anybody else