Last week, Binance founder CZ participated in the EASY Residency Season 3 closed-door workshop, where he shared his experience and insights with attending startup founders. He discussed various topics, from building the first team when establishing Binance in 2017 to forging cohesion under pressure. CZ candidly reviewed Binance's journey from 0 to 1, reflecting on early days in a shared office, personal challenges, and deep thoughts on privacy, AI, talent turnover, and global differences. He also openly shared mistakes made in hiring, decision-making, and transformation. Below is the translated transcript compiled by BlockBeats, with the conversation record provided by YZi Labs.

Interviewer: You founded Binance in 2017 and recently published a new book titled "Freedom of Money." In the book, you mention many early members. When building the first team, what qualities did you value in those people?

CZ: It actually requires many different qualities. I think the first point is: you need to get along with this person and trust them. For everyone, this sense of "harmony" is different. The second point is, you need to look for people with complementary skills. For example, He Yi was in charge of all backend tasks: company registration, finance, HR—she excelled in these areas. Roger had strong technical skills and was also a very strong marketer with strong execution ability. A complementary team is very important.

Then, you need people who are willing to work with you, and you need to earn their trust. As a founder, you need to cultivate and build a team that is essentially willing to follow you to the ends of the earth. This takes time to accumulate.

I think these are key qualities. Additionally, I work very hard, and everyone on the team works hard as well. When combined, it works very well.

Interviewer: How did you meet He Yi? What about other team members like Alan and Roger? How did you recruit them?

CZ: Everyone's story is different. Most of this is written in the book.

I met He Yi in a WeChat group. There were many cryptocurrency groups at that time, right? Later, we met at a conference, exchanged greetings, had some business discussions at another event. I knew He Yi for a longer time. She initially worked at my friend's winery, and later I learned she had a job at a bank. Then I recruited her for a previous startup project of mine. That project failed, so I recommended her to work on another startup project. Later on, I brought her back.

Roger was introduced through a mutual friend, while Alan actually "inherited" from someone else's team.

So everyone's background is different. But when you meet someone you can work with and genuinely like, you establish a connection, and it's a gradual process.

Interviewer: In your book, you also described the early days of Binance, your own state, such as sleeping in the office, losing 10 kilograms in three weeks. There is also a scene: an investor came to your office at 10 p.m. on a Friday night, and it was still bustling like a vegetable market inside. How did you create this kind of energy to make the early team willing to work with you?

CZ: I think the most fundamental point is: you have to lead by example. I don't know if there's any other way.

If the founder is not 24/7 committed — of course, you don't have to be 24 hours, maybe 10, 14, 16 hours a day, sometimes longer, sometimes shorter — but if the founder is not hustling hard, and you want to hire a bunch of hustlers, it's just not possible. This kind of atmosphere cannot be created.

I've done some entrepreneurial projects before, and I was part-time at that time. That's why I don't invest in part-time founders now. I tried that before, and some investors told me, "CZ, this doesn't work, you're part-time, the people you hire leave at five."

I think the key is: you have to lead by example. Then you will know if the rest of the team can keep up. In a small startup team, if everyone is hustling, things will move very fast. Bureaucracy should be minimal — in a large company with thousands of people, you need coordination between different departments, legal review, compliance review, like Binance today. But in a small startup, you can't have those things.

To be honest, I don't know how else to describe it. But most truly successful startups have a culture of hard work. There's no such thing as "working hours." Of course, you have to sleep, you can't always sleep in the office, but overall, you have to work very hard. Especially in today's environment, you are competing with people globally. For example, if you are doing a prediction market project, there are other teams doing the same thing. Unless you are much smarter than them, you have to work harder than them.

Interviewer: What is the state of a founding team under pressure? How did Binance unite and become stronger than before? There are some interesting psychological mechanisms at play here.

CZ: Teams usually do not collapse under pressure. Most of the time, teams collapse when there is nothing to do. If you work well with your team, the greater the pressure, the stronger the team's cohesion. The best team building is actually fighting together in the trenches. I don't know if this way of saying it is easy to understand.

There is an old movie called "Black Hawk Down," I don't know how many of you have seen it. There is a person in it going to war, and someone asks him, "Are you addicted to war? Why do you always risk your life? Is it for honor? Trophies? Or something else?" His answer is very classic: "When the first bullet flies past your ear, none of that matters anymore. What matters is the person next to you."

Humans have a natural psychological dependency on those who fight alongside them in the trenches. So, if you are fighting in the trenches with your team, it is actually the best team building.

In the early days of Binance, external pressure, especially political attacks that were not caused by ourselves, were actually the best team building activities. Going out to eat together or doing team building activities can't compare to this. Solving problems together is much better. These troubles did not cause us much trouble. We are a team that withstands pressure together.

The workload is indeed large, so physically, it can be tiring. Our CTO had inflamed eyes, this is a real physical reaction. But from my experience, I have never seen any team collapse under pressure, especially external pressure.

But if you have no revenue, no business, then that's another story. If you have users coming in, business coming in, the system under pressure, those are actually "good problems." If you don't even have these problems, it means your startup is not doing well.

Questioner: When you are not in meetings or not present, how do you keep the team's sense of urgency and product vision alive?

CZ: You need to set the tone, and then this will become a culture and expectation. This is actually saying: your work intensity depends on how much time you are willing to invest. As a founder, you need to have a very efficient way to judge how long something will take and you need to set that expectation. For example, in the early days, I would say, "There is a user complaint about this bug, can we fix it today?" The expectation is that we need to fix it quickly, we cannot do a too complex analysis.

For example, "Tomorrow we are going to switch to a different ICO method, we are going to allocate based on user deposits. This is different from what we did before, and we need to go live tomorrow." Everyone understands this must be done.

You set the expectation on time, and that determines how complex the work should be. I have always pursued a very fast turnaround time. I strongly recommend adopting this mindset. When you do this for a week, a month, everyone will get into the rhythm, and then you can maintain that culture. The problem is, as the team grows larger, the speed slows down. That's the advantage of a startup team.

Questioner: One more question about AI. With AI in place, how will you operate your business differently to make things faster?

CZ: I think AI is a tool for improving efficiency. I would set higher efficiency expectations—such as the speed of execution and the amount of work one person can do. However, AI does some things well and some things poorly. You need to understand that. Some things AI can help with, some it cannot. So for me, it hasn't fundamentally changed anything. AI is just an efficiency tool that makes people more productive. You just need to consider it and incorporate it into your expectations.

Questioner: How did you initially seize the market opportunity? How did you strategically find a key differentiator, then focus in that direction, capture the market, and grow? There were competitors at the time; how did you win?

CZ: To win, you need to do about a hundred things relatively well. And to be able to do a hundred things well, you probably need to try ten thousand things. Many things you do, try out, but they don't work well, so you need to adjust quickly and iterate. I might even say "pivot." Overall, you need to serve your users well. That's very important. On most platforms, you need a solid product, good security—especially in the crypto industry. Then you need to try many different things. Those that work, you double down and amplify.

For example, when we reduced fees, what actually happened was—we lowered fees by 50%, and the transaction volume increased X times. So we were in a good position to do that. Then we continued to amplify, lowering fees again. When we release a new feature and it doesn't perform well, no one uses it, we shut it down and move on to the next. For features that users like, we double down.

So there's a lot of trial and error. At least for me, my style is—it may sound a bit crazy—I don't really make detailed plans. Most things happen organically. I don't have a grand overarching plan.

In this day and age, it's difficult to predict what will succeed. Some features work, some don't. My style is to engage heavily with the community. I chat with everyone in the "trenches," so I can feel the feedback. I don't chat with everyone, but Twitter (now X) is a good channel to get feedback. Then you iterate based on that feedback.

But if you don't plan, you have to have good enough judgment to recognize what works and what doesn't. So you must have that feedback loop. Like a beginner martial artist, you'd plan for five steps, but no matter what the opponent does, you follow those five steps. But a true master knows how to react, knows how to respond. You just need the right response.

Questioner: And then how did you become a leader in the entire industry after that? Is it just a continuous iteration, continuously speeding up the product for users, and then eventually reaching that position?

CZ: Yes, I think it's all about continuous iteration. We started with a spot trading platform, then we added futures. Then we saw others doing DeFi, and we did that too. When the Meme coin craze really hit, Binance itself was dealing with the regulatory issues in the Biden administration era, so we were actually behind for a while. We were distracted, not paying attention to the market. But later, we quickly adjusted. By that time, we had one advantage: a large user base, so we could catch up.

But in the early days of building a user base, you just need one good feature that people are willing to use. You don't necessarily have to be the biggest in the industry, but you should strive to be the first or second in your niche.

To be honest, more people are starting to use decentralized exchanges today, so that space will grow. We will still be a significant player in the centralized exchange space, but maybe not the largest in the whole industry. And that's okay. You can only do so much.

So I think the goal is: first, have a sustainable, profitable business. Second, grow it to a scale you can handle, not too focused on competitors. Being the first is not the goal. Doing what you are best at, creating value, that is the goal.

Questioner: Binance has come a long way since 2017, going through different stages, from a small exchange platform to now an incumbent giant. With so many things to balance, how do you decide what the company should focus on at each stage? How do you make decisions?

CZ: For me, the core now is "people." I cannot focus on every aspect of the Binance ecosystem or Binance team. I'm just a shareholder; in fact, I no longer manage Binance. EZ Labs has its own team, BNB Chain has a group of core developers, and there are ecosystem projects, and so on.

So at this stage, what I mainly do is: find the best people and empower them. I am no longer the founder myself, no longer personally leading and doing specific things. But my advantage is that we have a lot of resources, reputation, user base, platform and ecosystem, as well as funds for investment, and so on. I am not being arrogant, but my current position is indeed different from that of a first-time entrepreneurial founder or a hands-on founder.

For most people, you still need to find the best team. For most people, you need to have a deep understanding of your business to know what the priorities are and what should or should not be done.

Overall, though, it's not difficult to judge what is hot and what is not today. You can see it. But there is a lot of noise in the industry. Predicting the market has been hot recently, and it was hot before that; before that, there was the Meme coin craze; and before that, the industry was very cold. As a founder, you need to be aware.

I don't know how others develop this kind of judgment. For me, it's simple—I spend some time on Twitter and I know. Others may look at data and so on. But if you only look at data, it may mean that you are sometimes out of touch with the community. You need to look at data, but as a founder, my approach is: you need to have "intuition" about the industry. You can't always rely on data. You can look at data, but you need to cultivate a sense of the market.

Questioner: That's very helpful. I'm sure Binance has gone through many decisions, right? Some of which have been great decisions. So, from all the mistakes you've made over the past few years, what is the most valuable lesson you've learned technically?

CZ: Ideally, you want to minimize the negative impact of incorrect decisions. When making a decision, you have to consider its impact. First, whether the impact of this business decision is large or small; second, on the flip side, if this decision is reversible, then even if the impact is significant, once you realize it's wrong, you can correct it midway. If it's a major irreversible decision, then you need to think more carefully.

For example, like flying to the U.S., that is a major decision that may be irreversible. But in product development, most decisions are reversible, right? You can launch a feature, and if it doesn't work, you remove it and add a different one. Some hiring decisions, while more difficult to undo, can still be done. Of course, you don't want someone to leave their current job to join your company, only to find out it's not a good fit two months later, and so on. So it is still reversible but very uncomfortable. If you are making a large investment in a field and it is a one-time thing, then you need to be more careful.

We have made many wrong decisions, but fortunately, most of the wrong decisions are relatively controllable, so you need to be mindful of this in major decisions. So far, I think we have done okay. Well, this is a relatively general answer.

Questioner: Thank you. What are some common failure patterns you have seen after a project gains early traction?

CZ: The first one is losing long-term vision. Some companies, once they start making money, have founders who expose different human traits. When a project starts making money, some individuals want to maximize gains quickly, so they take more benefits for themselves, often at the expense of the project's long-term outlook. Long-term money is usually much bigger, so that's the first point.

The second point is the growth of the team. Managing a team of 20 people is different from managing a team of 200 people, and managing a 2000-person team is entirely different. When you are dealing with a team of 20 people, you may only have a few new members, maybe no external investors yet, which is relatively simple. But when you grow to 500 people, with 10 different external investors, the CEO's job changes. This transition in roles can sometimes catch people off guard. So there are a lot of potential reasons in this aspect. A very promising startup can fail; I cannot list all the reasons, but this transitional period is crucial.

Going from 0 to 1 requires a specific set of skills, going from 1 to 10 requires slightly different skills, and going from 10 to 100 is different again. In Western countries like the United States, investors often force founders to change management, give up control, and so on. In Asia, for example, it is often the founder and CEO who need to lead the company all the way. The dynamics under different cultures have some differences, and there is no absolute right or wrong, but you must be very clear about this transition.

Questioner: I am a big fan of cryptocurrency and have been using it extensively in the financial field for a long time. My question is about the so-called binary opposition between the "Global South" and "Global North." You see, if you look at Crypto Twitter, or when we as investors observe the market, people often pay more attention to meme coins or more speculative things. But if you look again, they do not pay attention to projects like Tron or Ripple, nor do they focus on how the Global South uses cryptocurrency. For example, I have spoken to people from Venezuela, and their way of using cryptocurrency is entirely different. I would like to know what are your thoughts on how to approach this difference?

CZ: Alright, there are several different ways of thinking about it. First and foremost, you may or may not need to separate them. I think you should first look at what features your product wants to target, right? You can look at geographic demographics or functional demand populations. That is: Who would use my product for what features?

To be honest, when I was running Binance, I didn't think too much about it. I just thought—well, we have a trading platform, whoever wants to use it can use it, and then we developed futures, we developed savings, our fees are very low. Many of our products are even free, such as just holding cryptocurrency on Binance, right? We have a free wallet. When we make payments, these transactions also have very low fees, no problem. Although there are indeed differences between regions, I didn't think from that perspective at the time. I just saw low-income users, high-net-worth users, active traders, long-term holders, savers, and so on. I was more thinking from this perspective.

But if you are developing a new product, you may need to be more specific. When we did the first product, it was just for traders—if you are an active trader, Binance is a trading platform. But later on, we kept adding more features. And today, I think most founders, especially in the crypto space, should take a global view, because cryptocurrency has no borders, right? It allows you to reach a global market and a global user base, which is a huge advantage compared to traditional businesses. Traditional businesses usually start from one country and then have to expand to one country at a time. But in the cryptocurrency space, that's why Binance has been able to grow so rapidly. We are a cryptocurrency trading platform, and initially, we didn't have fiat channels, so anyone could deposit, anyone could withdraw on the blockchain, which is a huge advantage.

Depending on what your product is, you can look at geographic location, income level, advanced users or beginners, or different types of features. But I would see the whole world as one unit. If you are working on a crypto project now, almost every country will have your users. Even if you are working on something related to AI, AI is also global, right? That's how it is.

Interviewer: When you are building a product, especially those that are most disruptive, you often spend a lot of time convincing others who initially don't understand. How do you determine whether you truly have foresight or if you are just being stubborn?

CZ: You need to have a core user base, no matter how small, but you must have a few users tell you that your product is really great. If no one tells you this, then what you have created may not be useful, or you can't find your users. Ideally, you should have at least—I don't know the specific number—but I have always stayed in touch with about a dozen very hardcore users who will give you feedback. You must get very positive feedback from these people. If you can't find 10 or 12 people who really like your product, then you may not be building the product in the right direction. You don't need everyone to like your product, right?

So there can be a lot of people who hate it—'I will never use this'—but you need to have those few who tell you, 'Oh, this is great, I've been using it.' And you need to see them really using it continuously. If you have such users, then you are likely to find other similar people. In fact, you don't need a large number of users to use your product to be successful. Of course, more users are better, but some niche products also do very well. So, there is a balance here. But you do need some users to tell you that you are on the right track.

Questioner: You often talk about privacy, so I have a related question. I would like to hear your thoughts: Looking ahead to the future of cryptocurrency and privacy, at the infrastructure level, what is truly important to you? Is it more focused on user protection, regulatory compliance, or seamless integration with existing wallets and trading platforms? Following up on this question, do you think in the near future, privacy will become a default feature of mainstream wallets and trading platforms?

CZ: Interesting question. I believe privacy is a very, very important feature. I think currently blockchain has sacrificed too much in terms of privacy. The tricky part is that doing privacy well and making it easy to use is very difficult. There are many attempts now, such as Monero, ZCash, and a bunch of other things, but so far, there hasn't been a solution that can achieve both very strong privacy and ease of use. So I think if you can address these issues, if you can find the right balance in these areas and push the privacy level higher, then there is a high market demand. Bitcoin is too easily traceable, too public. That's my view.

Questioner: How important do you think privacy coins should be in terms of handling compliance?

CZ: I'm not a compliance expert. I try not to answer compliance questions because I think it varies by region. I am indeed surprised that Monero has not done bigger. But Monero is difficult to use, slow, you have to download a fairly large blockchain, and apart from holding the coin, there is nothing else on the chain. No smart contracts, no exchanges, no other functions. So, if you can build an exchange mechanism default with a privacy feature, I think there will be many different types of privacy solutions in the market. I don't know which one will prevail because I can't predict, and I can't predict user psychology well, so you have to try, figure out which one will succeed, try different things with multiple projects, hoping some of them will succeed.

Questioner: We just talked a bit about the different talent needs at various stages in the field of decision science. My question revolves around building your team: at different stages, what are some of the most correct things you have done to find the right talent? And how do you maintain that relentless culture within the organization?

CZ: Well, to be honest with you, I've made a lot of mistakes, especially in hiring the wrong people—not that they were bad people, but I put the wrong people in the wrong positions or had them do jobs that didn't suit them, right? So sometimes I hired really good people, but sometimes I hired people with too much experience, or I hired people who didn't fit their roles, and the result wasn't that great, right? And then we have to (adjust). I do believe that as a founder, you need to constantly experiment, you need to do a lot of hiring and firing—especially the firing part which is not pleasant, but it has to be done. I do suggest you should maintain a relatively high turnover rate. I think the industry standard for the tech sector is 20% per year, and you should at least aim for that level.

However, in my team, we usually look at the data. Those who joined early, once they settle in, become very loyal. In fact, I was too loyal to old employees, which in turn left less room for new people to grow. In addition, most of Binance's founding team members are still here, which has its pros and cons. The advantage is that they have a long history, team stability, and so on; the downside is that not every founding member can grow at the same speed as the company, especially for a rapidly growing company like Binance, or unicorn startups of this era. So sometimes you have to rearrange the positions of some founding members, and at times that may include yourself. I have involuntarily changed positions myself, but I think it was a good rotation.

So I think founders need to be very active in team attrition, both voluntarily and involuntarily. Frankly, from a numerical perspective, I think this ratio should be between 20% and 30%, depending on the company and the situation. Yes, you will make mistakes, but I would say it's more about being "unsuitable" rather than "ill-intentioned." I just put the wrong person in the wrong place or put the right person in the wrong place. Especially in areas I'm not very good at, like legal and compliance, backend and legal compliance, I didn't have the right people at the right time, which caused me a lot of pain later on and led to many problems. So I do think you need to rotate personnel frequently to find the best match. Yes, there is never perfection.

Interviewer: You talked about leading by example and working hard. How have you maintained long-term discipline and perseverance?

CZ: This question is straightforward. You must be passionate about what you do. In other words, you must enjoy what you do, what you build must make you happy, right? If this is not the case, you cannot work so hard. If you feel like "I'm just working hard for money," it doesn't work, and the people around you can feel it, right? Usually, that doesn't work. The goal of a startup should be a bit more than just "making a lot of money" and a bit less than "saving the world," somewhere in between. For us, we are working to increase financial freedom for people worldwide. It's not entirely about saving the world, but what we are doing is quite meaningful to you. You must be passionate about what you are building; it must be your mission.

Questioner: My question is, when was the first time you realized you had the ability to build such a massive company like Binance? Did you believe this before we started, or did you only realize it after Binance began operating? Was there a specific turning point that made you think, "Well, I can really make this happen on a global scale"?

CZ: Yes, but I initially had a very international user base because that's how cryptocurrency and cryptocurrency exchanges are. So people from all over the world came to us. To be honest, at different stages, I was not sure if Binance could always stay ahead, even today, right? Competition always exists, and you never know. For me, I just tried my best and saw what would happen. Again, my goal is not to become the world's largest company or the world's largest cryptocurrency company; that is not the goal. The goal is just to do what we are doing really well.