This is an excerpt from a session with Alchemy Co-founder CTO & Mayfield MD Navin Chaddha at TechCrunch Early Stage 2022. The full recording and transcript can be found here.
Navin Chaddha:
Pleasure to have all of you here. I have the great privilege of having Joe Lau, Co-founder and CTO of Alchemy with me. Today, we are going to be talking about how a company goes from zero to leading Web3 blockchain platform company.
So Joe, let’s start with, what does Alchemy do? It’s been one of the fastest growing successes, from 0 to $10 billion in market cap in two and a half years.
Joe Lau:
So yeah, at a high level, Alchemy is a blockchain developer platform – think of us as the Amazon Web Services for blockchain. So in the same way that AWS helps companies build internet and software companies, we help people build blockchain applications more easily and more effectively. And breaking that down a little bit, we see three big shifts in technology. There was the personal computer, there’s the internet, and now there’s blockchain. They each add a new type of primitive and a new way to build an application that you couldn’t have before.
And the really interesting thing is, if you look back in history, what you’ll see is for each of these technological shifts, there’s always a developer platform that pushes that technology forward. So with the personal computer, it was Microsoft Windows, Mac OS – the operating system is what let people build applications like Microsoft Word, Chrome, et cetera, and those things are what brought value to people’s lives. With the internet, you saw the same thing with Amazon Web Services, which enabled a whole slew of companies to be built.
We thought for blockchain to be successful in the same way that these other technological shifts were successful, someone had to build that developer platform, and that ended up being us. So we started the company in 2017, today we power about $100 billion in transaction volume annually, and millions of users all around the world.
Navin Chaddha:
That’s great. So how did you get the inspiration? Can you talk about your founding journey with your co-founder, and why you chose blockchain this versus so many other things?
Joe Lau:
I met my co-founder Nikil in college, we were TA-ing a databases course. Most college courses as you know, are 100, 200 people, something like that, ours that year was 150,000 people – so just a little bigger than your normal college course. And the reason for that was, it turned out we were piloting this revolutionary new concept at the time called massive open online course. But we piloted essentially what would eventually become Coursera, and we were basically thrown into that environment. So from the very beginning, when I got to meet my co-founder, it was almost like a startup environment, and that’s kind of how we got to know each other.
When we graduated, we were really thinking about what we wanted to do with our lives. And probably like a lot of the audience, the thing that was really important for us was impact – how do you think about how to leave as big of a mark as you can on the world. And for us, we measured impact on a couple different axes. The first axis was, how many people’s lives you could touch, the X axis, the Y axis was how deeply you could affect people’s lives, and the area under that curve is essentially your impact. That’s how I thought about it. The X axis is pretty clear. Again, being a software engineer, how you affect life on a billion person scale is with software. And the Y axis, originally we thought was actually not blockchain, because this was before blockchain, it was more of a personal thing at the time.
In the very beginning, what we cared about was how to bring people closer together. We had graduated from college, we felt distant from our friends, my co-founder was going through a rough breakup and a lot of my friends had moved away. So we were building consumer social experiences and applications to help recreate that sense of community.
We did that for a couple years, 2014 to 2017, and nothing worked. We tried like 9 or 10 things, it’s really hard. I think the 10th or 11th thing we built started to take off – it was doing well, it was on top of the app store and stuff, but at that point, we really started to see blockchain take off. So 2017 was when we really started to see Ethereum take off. And the really interesting thing was, it went from just Bitcoin being a currency or store of value, to now Ethereum being a true and complete platform that you could build all different kinds of applications on top of.
And that to us was a really revolutionary shift. Now thinking back to those three shifts, the personal computer, the internet, blockchain, first you had the personal computer which added a new primitive by allowing machines to follow human instructions. Then you had the internet, which adds another primitive, machines can exchange information with each other, and then now blockchain finally, you could have machines exchange value with each other.
So you could build new types of applications that you could never build before, and we got really excited about that. We jumped into this space and started building. Long story short, we found out it was really, really hard, so I ended up having to build tools and infrastructure for ourselves. And then over time, we would talk to friends and find out they had exactly the same problems that we had. So we launched our infrastructure and our tools as the platform that anyone could use, and was off to the races from there.
Navin Chaddha:
Do you think that’s how you got product market fit or did you have more work to do? Because that’s the hardest thing for a startup – everybody has an idea, but how do you make sure the product you build is a painkiller, that it is solving a real market need? Did you guys talk to users, what was that journey like?
Joe Lau:
We spent a lot of time looking for product market fit. For those who are past it, it’s probably the hardest thing you’ve ever done. For those who haven’t done it yet, I understand your pain, and it’s something that takes a lot of time. I think for us, we spent a lot of time looking for product market fit, and really finding the right thing.
I’d say a couple of thoughts there. I think one thing that helped us was really, really understanding how to talk to users. So for those of you who haven’t heard of this book, The Mom Test, you should go buy it right now, it’s an amazing book. It talks about how to talk to users and really understand their needs. The problem with users is they’re all like their mom, and that if you go to them and you ask, “Hey, will you use my app for this”, they’re always going to say yes, because they always want to tell you what you want to hear. And the reality is, after they sign up, do they actually use what your platform is for? Not always. So how do you tease out whether someone will actually use what you build, and how do you figure out if what you’re selling is actually what they care about? So that was a big learning for us.
I think the second thing was figuring out how to really speed up our iteration – this was one of the biggest learnings that we had. At the time, we were building iPhone applications and we would spend a week building something and then ship it off to the App store. At the time, it was a one to two week review process, so it’ll be on the App store three weeks after you originally made that change, and then by the fourth week, your users have mostly updated. So it’s really a month that you’re looking at to test one iteration. But when you’re looking for product market fit, the thing that’s most important is how fast you iterate. It’s kind of like machine learning and gradient descent, it’s about taking as many steps as you can, as quickly as you can, to find that local minimum, or for product market fit, that local maximum.
So for us, what we did was we really worked on tightening that iteration cycle, so we took that month and we actually dialed it back down to minutes. We ended up taking our apps, and instead of shipping it to the App store, we’d go to Berkeley and we’d pretend to be students. So we talked to students and we’d show them our app, and then when they ran into issues, we would basically get on our computers, change it right there into a new experience, and then show the next set of users. So we were able to dial that month-long iteration cycle back down into a couple of minutes, and that was how we were able to iterate super, super quickly.
The last thing I’ll add really quickly is, talking to users in the flesh, in real life, is really, really important. One thing we did for the social consumer stuff was put our phone numbers directly into the app. It ended up being a bit of a misplay, because we got millions of users, and turns out your phone doesn’t work after 15,000 texts. That’s something that we discovered, so don’t do that. But the important thing is, you want to talk to your users.
For the blockchain stuff, we went to conferences and we talked to everybody that we could talk to. And it was hard, we got a ton of nos in the beginning. You’re talking to people, you’re trying to get them to use your product, you think it’s amazing and everyone is like, I don’t care. You’ve really got to be okay with getting those nos and just getting the user feedback that you can at the time. So if you are getting nos, just know that’s part of the path to finding product market fit. That is totally 100% okay, and the important thing is pushing through that and eventually figuring out what people want.
And that was what we did. And there’s always a bit of hindsight is 20/20 or whatever, but that’s what worked for us.
Navin Chaddha:
So a lot of times when startups get founded, they either go after an established market and you have a faster, better, cheaper product, or they go after an emerging market, what is often called a blue ocean opportunity. So what were the lessons, if any, in navigating a new market? Because when you started the term Web3 wasn’t coined, so if somebody asked you what’s the market size for a platform company, it was probably at zero. So how do you end up navigating this?
Joe Lau:
Yeah, that’s a fantastic question. And I think one thing is, in the very beginning, in a lot of cases, the blue ocean is more of a blue pond at first. When we first started on our blockchain journey, Nikil and I were raising one of our first rounds and when you’re talking to investors, you want your TAM to be as big as possible. So we’re using the most crazy estimations of how many people, how many teams were in the space, and how much they pay. I think we got to 10 million for a total TAM, which if you’ve ever made a slide deck, is not an inspiring TAM. And we were like, how do we make this figure?
So I think at the very beginning, emerging markets are going to look super small. The numbers aren’t going to show at all, you have to have a thesis or you have to have some reason. For us, that reason was a couple of things. One, we saw the growth was really fast. The second thing is, we saw something that I always think is a leading indicator for real technological progress, and that is we saw a lot of our smartest friends and coworkers jumping into the space on their nights and evenings, and some of them eventually even in their day jobs. And that’s always a leading indicator, because you have to have the builders before you see any progress. But we started to see an acceleration of that, started to see that excitement, and thought, hey, maybe there’s something really interesting here.
And I think if you’re going after an emerging market, you have to be okay with not having the guarantee that the market’s always going to be there, which again, stepping back is totally fine. It’s either going to be, there is no big untapped market, there’s either a small untapped market or there’s a big market with competition, that’s just kind of how it works, so pick your poison.
If you go after the small untapped market, hoping that it’ll grow, I think one really important thing is, it changes quickly and you have to be talking to users even more, because user needs are changing all the time. So again, in the very, very beginning, every single one of our customers would have a direct chat with us. We didn’t have customer support people, we had engineers actually on the phone and on chats with people. So we could constantly get that iteration cycle, because when customers ask for help, one, they’re asking for help, but two, they’re actually telling you what to build and they’re telling you what problems they want solved. A lot of companies will have customer support or someone else handle that. And what actually happens is that the most valuable product feedback never actually makes it back to the product team, and it never actually gets built back into the product. So that’s something that was really helpful for us.
And I think at a high level, I’d say – be okay with it being a blue pond in the beginning, but at least have a thesis for why you think it’ll eventually be a blue ocean, and then figure out how to stay on top of that wave and learn from users as quickly as possible, because the space will change really quickly. And for those who know blockchain, the blockchain space has radically changed in the last couple years, and the only reason we were able to keep up was because we’re always talking to the users every day.
Navin Chaddha:
Yeah, I think persistence and perseverance just do magic for startups and founders. So you briefly touched on you and your co-founder living together, working together, TA-ing together for a 100 student class that became 150,000 student class – anything we can learn? How do you choose a co-founder, how do you work with one another, because you’re not going to agree all the time? And it seems for you guys, this has been a great working relationship and a great partnership, and I feel you guys complete each other’s words, and if one is there, he’s always talking about the other, and you still make joint decisions.
So anything we can learn about how do you pick a co-founder, how do you work together, and then how do you make tough decisions?
Joe Lau:
Yeah. I mean, the first thing is, get a co-founder. Who in here is a solo founder? All right. We’ve got some solo founders, all the more respect to you. For us, what we found is, it is tough being a solo founder. I think when you have someone who’s right there with you and there through the ups and the downs, what happens is you’ll end up balancing each other out. Sometimes he’ll be having a good day, I’ll be having a bad day and my co-founder will help push us forward and vice versa. I think you have that like self equalizing and self-regulation mechanism that’s really, really helpful. Honestly, my co-founder is awesome. I think working with my co-founder has been one of the best decisions that I’ve ever made, especially when it comes to the company, I know he’d feel the same way. So that’s the first thing I’d say, is highly, highly recommend, if that’s something that you’re thinking about and you’re not sure.
The second thing I’d say is, I think we got really lucky. I think you want to take your time picking an awesome co-founder. My co-founder and I are friends, so we met in school. We lived together for five plus years, for those five years, I saw him more than my girlfriend, now fiance – we were practically married. And to your point, there are going to be hard times. There are going to be times when people leave, there are going to be times where you disagree. I think we’ve been really lucky in that we’ve never had a fight, we’ve never had a huge disagreement or anything like that.
And I think part of that is because one, we had that base of friendship, and two, we shared a lot of similar values for how we approach the world. And there’s that concept that if you take two smart people and you give them the same information, they’ll generally reach the same decision. And people’s decision making processes are based on biases and priors and other things, but if you give people the same information, a lot of times you can reach an agreement, and that’s something that we’ve always lived by. I think that’s at least what’s worked out for us.
Navin Chaddha:
I think we’re almost out of time – thank you Joe for joining us.