I remember people saying many times in our San Francisco office, "Wow, it's so refreshing that you and Matthew still show up for work."
When did showing up become optional?
I'm Jimmy Allen, and this is "Founder's Mentality: The CEO Sessions."
Michelle Zatlyn co-founded Cloudflare in 2009 with her business school classmates, Matthew Prince and Lee Holloway. Today, Cloudflare operates in 125 countries and protects 20% of the world's internet traffic. Of the three founders, Michelle was the one without an engineering background, which is what I thought this conversation was gonna be about.
I would love to start with just the very simple decision to be a co-founder of Cloudflare and what that was like as a non-tech person to be part of the starting of a tech company.
I had two great co-founders, Matthew and Lee, and we were really excited about what we were working on. And while I wasn't an expert and it wasn't like a clear slam dunk, sometimes people start companies in something that they worked in before or knew a ton about, and that's great. Like, those are assets. Actually, a lot of people who start companies also do it as an outsider, and that's actually also a great asset because as long as you have a high rate of learning, you get to question everything. Almost kind of like from a beginner's mindset of saying, hey, I'm not already an expert. I don't have a point of view on this. Let me go and run this down. Of course, I was like, "Do I even have the skillset to do this?" And all these different things. But I think as I look back, thank God I didn't listen to any of those notes, because the reality is like, no, of course I didn't have all the perfect skillset. But you know what? No founder does. And it's putting yourself in the situation, having to make decisions, having to learn from those decisions, and how fast you can learn and then make another good decision actually is what makes you a great leader or founder. It's not what you go to the table with.
You started with an insight that very few single founders start, which is no founder has all the skills necessary to start and build a company. It's always a founding team. And most discover that only after a couple crashes and burns. It feels like that's the one insight you guys did have, that we have complementary skills and will somehow figure it out.
I think starting a company is hard and actually like the stats are most companies fail, and it's hard to actually solve a problem that's meaningful. It sounds so easy and trite, but actually it's hard to solve. Go find a problem that a customer needs solving that actually you need to do more than once to build a business around it. And so early on, before Cloudflare was born, Matthew and Lee had another initiative called Project Honey Pot. This was a community that cared a lot about cybersecurity, and so we were surveying them to understand more about the problem set when I got some survey results back, market research back from this group. And it was kind of like, tell me your pain point around cyber threats online. And again, this was back in 2009. It was a very different world than it is today. Some of the qualitative answers were the best part, Jimmy. It was things like cyber threats are the scourge of the internet. They're criminals and they steal resources. It was a very visceral response from this potential customer, kind of our ideal customer. There was other ones like cyber criminals make me believe in the death penalty. These were not casual answers. They were written in a way that when you read them, they kind of jumped off the page. And so it was clear there was a real problem here.
And when we were starting, like any eager founder would, I was reading a ton. I read just like anything, you're gonna start something in an area, you're not an expert, you're trying to read a ton. And I was reading books about founding, and it was interesting. I remember this one book I was reading at night, and I remember the sentence, "95% of startups fail." And I took the book and I threw it off my bed and I was like, "Why am I doing this?" So I was like, "Okay, we can't be one of those 95%. I don't wanna throw away this opportunity." And so then we became so obsessed of, are we actually solving a problem that people care about, that our customer cares about?
What was that like as a non-tech person starting a tech company? The phrase keeps coming up to me, which is kind of customer-led tech, which is that it's just not true you're a tech company, you're a customer company that serves tech and that you feel like you have a good feel for customer. Is that fair? Is that a focus?
It was interesting that early on, it wasn't data that actually was important to us. It was actually some of the qualitative feedback we got from our early customers that we were doing Skype calls with every time someone signed up, and after the fact of saying, "Because of Cloudflare, this is the first time in five years I've slept through the night." And honestly, it was like those were the things is what we recruited on, we raised money on because they were so emotional. And I think that if you can solve a real problem, a pain point for a company or a customer, there's no bigger responsibility, and it just feels incredibly empowering. You're like, "We built something that actually solves a problem and delivers real value to somebody." Then you're onto something. And that seems so obvious, but some people start companies that actually don't solve real problems or they solve something that 10 other companies already solve. And you're like, "Okay, but do we need an 11th solution in the market?" And maybe we do, maybe it's better, but are you sure? And how do you know? And you have to hold yourself to a high test. Even to this day, I love spending time with our customers. Our whole executive team loves spending time with our customers because when you talk to the customers, all of a sudden it's very clear what the priorities should be. And in every single growth company, there is more to do than time and people to do it. And so getting the order of priorities right is really important. And one way to do that is by talking to customers.
And the issue of listening to customers, I mean, is that been something that you've been able to provide is kind of get into the tech conversations and say, look, I can at least put the priorities down as our customers see them.
Well, it's interesting. I'm just listening to your reflection. I started a company in an area that I was not an expert in. I was technical in a different sort of way, but I wasn't technical in this space. And so there was a lot of learning. I was drinking from a fire hose when we started Cloudflare. And the reality is, today I am an expert. And it's because I've showed up every single day, and I've learned a ton.
Guys, I need to make a mea culpa here. I walked into this conversation with a pretty strong premise, non-technical founder, room full of engineers. How the hell did she survive? Well, Michelle shut that down in about 30 seconds, and she's right. I mean, Cloudflare is 16 years old. She's been doing this every day for 16 years. Of course she's an expert. The more interesting question is how she became one.
When we started Cloudflare, there was eight of us. We worked in an office in Palo Alto on Emerson Avenue. We used to go to lunch every single day together and we had basically all engineers and me at this company. And everybody was technical, checking in code except for me. And we were talking about technical things all day over lunch. We were building, again, back to sorts of things. And I'd go back home, back to the office and I'd make a list of everything I heard and I'd go and I'd look it all up. And so kind of by the end of the week, now I was talking a lot about reverse proxies and DNS and all these sorts of things. And if you have the courage to put yourself in that situation where you feel like, I don't know everything. And that's hard, but that's how you learn. Now we have an amazing customer support team, but the first two years of Cloudflare, Matthew and I were pretty much tied for the most number of support tickets answered. And you learn a lot because those were technical questions. And I loved that and I loved learning more and I loved hearing people even more ahead of me in the technical ability so that the next time I could do it myself. I'm competitive. I wanna win. I wanna be able to do everything myself. I don't wanna have to call an expert on something. I wanna be like explain it myself. And thank goodness I did that.
Most CEOs I talk to today when they finally kind of confess will say they feel like they're being held hostage by the tech, made much worse by AI, that they don't understand it. There's big budgets for it. People are coming in telling them they have to do X and Y. They don't have the instincts to prioritize. You said, well, one thing that a CEO has is a beginner's mindset, like break it down into language that you understand. The second thing is you said that, yeah, but customers set the tech priorities.
If you are a C-level executive and you feel like, I don't understand all this tech or AI, but I don't know where to start, the best thing you can do is just put yourself in the situation and then go and try and connect the dots after the fact. And then by the fifth time, you're gonna know a lot more than the first time. No one will remember the first four times that you didn't know the right thing to say or you weren't sure. But gosh, by the fifth time you start to feel better, then you start to contribute more and then you start to get the feedback loop. And all of a sudden you're like, I do know more. The worst thing you can do is not put yourself in that situation 'cause you don't want to look unknowledgeable because then what happens is you really are unknowledgeable. And all of a sudden you don't know the questions to ask, you don't have the right instincts, you haven't had the four mishaps to get to the good one, the rate of learning.
When we got to like a 350 people, we started to attract talent from what I would say like Silicon Valley talent who had worked at other founder-led businesses. So we were really excited. And I remember a couple people saying many times in our San Francisco office, "Wow, it's so refreshing that you and Matthew still show up for work." And I didn't really understand what they meant. And finally I was like, "What are you talking about?" And they're like, "Well, my last founder stopped showing up for work." I think it's because founders at the scale and you start to get real revenue and you have to learn about finances and governance. You don't know what to do, so you just stop showing up to those conversations because you don't wanna look like the unknowledgeable person in the room, but then you aren't a good business leader 'cause you're not in the room. You might know the tech, but you're not a good business leader. And so I think like one of the best things that Matthew and I have done over the course of Cloudflare is we keep showing up for work. And even when we're not experts, we show up, we surround ourselves, we ask the questions and we learn so that you have better instincts. You gotta stay in the field, you gotta stay in the game. Otherwise, it's just so easy to drift. And I think that's true for both executives with technology and AI, and it's also true for founders or employees of high growth companies who wanna get to the next stage.
You know, I spend my life talking about how companies go from insurgents to incumbents to struggling bureaucracies. And we ask ourselves, why do leaders let this happen? For older companies, the answer is often this, because the current leaders were never around when the company was an insurgent. They joined a big complex firm, and they've learned to lead an incumbent mired in complexity. But why do founders let it happen to their companies? They know what it's like to lead a small startup. Why did they let bad things happen? Michelle hints at an answer. Maybe they just stopped showing up, stop leaning in, stop learning. We know this. As the company scales, the conversations change, finance, governance, new functions, different meetings, and many founders just choose not to be in the room where it happens 'cause they don't like those rooms. They don't feel they can add value to all these new ways of working. They opt out, and suddenly they're not running the company anymore. They're just watching it. This applies to all of us. Increasingly, leaders are being asked to lead through things they don't fully understand. AI is the latest example. And when that happens, the temptation is to defer to the experts to not show up in every room where it happens, whatever it is. But we wait outside, we try to learn enough to participate again in key decisions and key meetings, but Michelle's warning you, you don't become an expert by staying out, you show up. And Michelle showed up over and over again until she was indeed an expert. So stay in the game, stay in the field.
Michelle, let's switch to the hyperscaling moments. And what I mean is you've used from 8 to 350, from 350 to 3000. How did you keep the culture? How did you keep your focus on what needed to happen as you became bigger and bigger and complexity creeps in, energy vampires creep in? How did you guys either adjust to the new scale or try to keep that founding culture of eight people, two of which are leading in customer support calls?
