The following interview is a conversation we had with Michel Tricot, CEO of Airbyte, on our podcast Category Visionaries. You can view the full episode here: Michel Tricot, CEO of Airbyte: $181 Million Raised to Build the Future of Data Integration
Michel Tricot
Thank you, Brad, for having me.
Brett
Yeah, no problem. So before we begin talking about what you’re building there, can we just start with a quick summary of who you are and a bit more about your background?
Michel Tricot
Yeah, of course. So I’ve always worked in the data industry. I started my career in 2007, been through different types of data, financial data, so medium scale, Internet scale data when I was at Library. That’s actually when I moved in the US in 2011. And then more like IoT, self driving, car type of data. And, yes, it’s utilizing all this experience I’ve built over the past 15 years to actually build air bytes and resolve a problem that is very real for companies, which is how they access data. And so I started Airbytes in early 2020 with my co founder, John and, yeah, we’ve been at it since then, and we’ve had a pretty interesting journey there. Originally, I come from France, but, yeah, I think at that point of my life, after living here for almost 1213 years, yeah, I can call San Francisco.
Brett
My home and a few things just to unpack there. What is it about the data industry that excited you to jump into it way back in 2007? And I’m sure the industry has changed a lot since then, right?
Michel Tricot
Yeah, I would say maybe it becomes a bit of a romantic story at that point. Just from as far as I can remember, even as a kid, I’ve always had this habit of collecting things, whether it’s books, whether it’s movies, whether it’s games, or whether it’s just cool websites on the Internet. And if you look at it, this is basically like holding data. And that’s always something that’s been with me. I like this idea that you have all this piece of raw data, and then by collecting it, you can start understanding what is the bigger picture behind this data. And you can start being smarter about what data you need to collect, where you need to collect it, how you need to access, and whether it’s luck that my first job out of college was in Beta or not. I think I was always drawn to it, but that’s the romantic version of the story.
Brett
Hey, come on. We like the romantic version. What about moving from France? How old were you when you made the move?
Michel Tricot
It was in 2011, I was about 27. 27 actually. I was 27.
Brett
Okay, so you weren’t too young then. So what was going through your mind at that point? I’m guessing you had started to establish a career at that point. So it wasn’t like you were a 19 year old kid just moving to the US.
Michel Tricot
No, but actually I’ve always wanted to come in San Francisco, in the Bay Area. I’ve always been into technology and for me it was this is where most of the technology is being created, is being innovated. Not to say that there is no other places where innovation is happening, but that definitely a place where you have a huge concentration. And as someone who’s passionate with technology and with data, that felt like a good place to be. And since I can remember, I’ve always wanted to move. I had a little I did my internship at the end of college. It was in 2007. I was more in around Princeton, New Jersey, working on medical imaging and medical data. But that was just for a year and a half. Went back to Paris and then finally went to San Francisco. Nice.
Brett
That’s awesome. And as I’m sure you’ve read in the media and you’ll probably hear from other people, there’s a lot of noise out there today about the decline of Silicon Valley and San Francisco and the exodus of investors and entrepreneurs leaving Silicon Valley in San Francisco. Do you think that’s true? Is that what you’re seeing from your perspective and what made you decide to stay here in the Bay Area?
Michel Tricot
That’s a good question. I would say every time something goes down, you have a lot of people who talk about how everything is actually going to be destroyed, not exist anymore. I feel like what is happening is that there’s been a lot of extremely high growth over the past ten years. And actually when COVID happened, it made this growth even stronger. And you can call it a bubble, you can call it whatever you want. There is always a point where the reality is catching up. But we should not discard that things that are being created are real. I’m just taking the Air byte example. Companies need to use data company to remain competitive. They need to have data. So the software that is being created is software that brings value. And today, whether we want it or not, there is still a huge pool of talent in the Bay Area.
Michel Tricot
I mean, in many other cities in the world, but there is still a huge pool of talent, there is still some capital. Just people don’t know really what’s going to happen with the economy. So maybe they are just taking a step back, trying to see what’s going to happen. And maybe they’re a little bit doom and gloom, but people are still creating value and from then I wouldn’t be super semistic nice.
Brett
That’s a super nice perspective to hear. Now let’s dive deeper into a couple of questions that we like to ask just to better understand what makes you tick. So first one, what CEO do you admire the most and what do you admire about them?
Michel Tricot
I really like Bill Gates to be clear. It’s someone who has gone from being an extremely technical, talented person and managed to basically create one of the most valuable company in the world. So building something from nothing and getting into most computers in the world and just the personality of Bill Gates and I’m a technical person and today I’m building a company. So in a way I’m trying to see, okay, what are the things and when did in his career change the trajectory of him as an individual to go from writing code in basic in a small office to becoming the CEO of one of the most powerful company in the world? Nice.
Brett
That’s a great call out. His evolution has been pretty wild to watch. There’s some of these old videos on YouTube when he was in his lawsuits back in when was like the he was ruthless, he was very aggressive, it was combative. It was a very different Bill Gates than the man that you see on TV today. So it’s cool to watch his transformation there. Do you take weeks off like him for what does he call him? His think weeks.
Michel Tricot
So I have a family, so I have to take some weeks off for sure. Also one thing in the startup is you work extremely hard and you want to make sure that also people at the company take the time to refresh, to think about what is the next thing they want to do. Like innovation comes when you have time to think. So I take weeks off for myself, but also as an example for people at the company that it is okay to take vacation. You work a lot during the week, you work a lot during every single day. You need to take some rest at some point. Yes. It allows your brain to just process a lot of things that happen and to just have the time to look at what’s coming next.
Brett
Absolutely. What about books? Is there a specific book that’s had a major impact on you? And this can be a business book or it can also just be a personal book that influenced how you view the world?
Michel Tricot
I would say there is one book that really changed how it was when I made the change in my career between being full individual contributor, full practitioner. At the time I was at a company called LiveRamp and I started to manage back in the day. It was a very small team and was just fortunate to go through the hypergrowth of a company going from scrappy startup to IPO. And I had to figure out what does it mean to actually, at the time, it was management, but what does it mean to actually lead people? And my CTO at the time just gave me that book from Andy Grove called High Output Management, and it’s written by the former CEO of intel. And although it’s very focused on manufacturing of microprocessor and some of the examples this book just made me understand, what does it actually mean to lead a team?
Michel Tricot
How do you evaluate yourself as a team lead as someone with managing and someone who is here to really build leverage with the team? That was a transformative book for me because, especially when you come from IC, it’s a different job. It’s not a promotion. You don’t get promoted into a manager. It’s just a fork in your carry and understand what are the things that you’ve been doing in the past? How do they change now that you’re a manager? How do you evaluate yourself and, yeah, that was a transformative book for me the moment I really embraced this idea that it was not the same job anymore.
Brett
Yeah, it’s such a great book and a lot of guests that come on, they share that as their favorite book as well. So I’ve read it. I will have to say, it’s not the most fun and entertaining book. It’s really meant to be a tactical guide to show you how to be a manager and a leader. I wouldn’t say that you just can’t put it down because you’re so excited and entertained every time you turn the page.
Michel Tricot
I was very excited when I wasn’t ready to get but no, to be clear, it’s also a book that every time I find what I would call rising star at companies or that I meet, it’s always the book I send. If you look at my Amazon order list, it’s probably the book I’m ordering the most and just sending it to people.
Brett
Yeah, I first heard of that book when I was reading an article in Fortune about Brian Chesky from Airbnb, and he had some learning approach called I think it was like going to the source or something like that. So he would just find the number one book that exists for whatever problem he was trying to solve, and then he would just read that book and master it. And that was the book that he had listed from Andy Grove. So it’s a good call out and great book. Now let’s switch gears and let’s dive a bit deeper into Airbite. So take me back to day one. Where did the company really come from? What’s the story behind the company? And then we’ll dive a bit deeper into the product.
Michel Tricot
Yeah, so John and I met in 2013 or 2012, and along the year we tried a few site projects together. But he was in his own setup. I was at Playground in Hypergrowth, so it was very complex to both be amazing at what were doing every single day. Plus at the same time having this side project. So we had a few of these projects, I would not call them successful, but actually what happened is we learned how to work together and we actually also build that desire to be working together at some point. And in 2019, I actually left the company I was at and I really started to go on this journey of, okay, now I’m going to start a company. And I started to brainstorm by myself. I actually took a month of vacation before starting it, but I started to really brainstorm.
Michel Tricot
And at some point I needed to have someone to run idea by. And I asked John if he was interested in just brainstorming with me. Maybe there was an agenda behind, but definitely I needed to have someone to bounce ideas. And very quickly as we’re going through this brainstorm, it became clear that we had both built a pretty strong experience in the challenges of integrating data and making data accessible. And at the end of 2019, we actually applied to YC. We did not really have a product at the time, but we had a problem space that were very familiar with. So we applied to YC with that a very early idea. And yeah, January 2020 was really when we started Airbite back in the day, it was called Dexterity. It’s a terrible name, but yeah, went through YC, and our goal was really, we need to learn as much as possible about this problem space more than what we know.
Michel Tricot
And we just spend all of our day, 910 hours a day, just on the phone, talking to data teams, talking to data engineers, talking to marketing teams. We built an initial product at the time, but that was very dear to our marketing. At the time COVID happened, marketing was not the best field to be building a product in. So we pivoted in the first six months of 2020, and it’s really in July 2020 that we set all on the idea of Airbag, which is how do we break down companies data silos and how do we make it in a way that is extensible, that is future proof. And that’s one of the reason why went for open source as one of the core thing of our product. And from then we just released Airbytes end of 2000 and 22,021, focused a lot on the first version of the product and building the community around Airbyte.
Michel Tricot
And we also did the fundraising that you mentioned before. And 2022 was just the product is becoming more mature. We’re starting to think also about more repeatable monetization of Airbite. And now in 2023, we have a team of 100 people and we have a pretty good product, very strong community. I think we have about 90,000 deployment of Airbytes. And of this 90,000 deployment, we have 3000 companies that are actually using Airbyte on a daily basis within their infrastructure. So, pretty strong achievement for the past two years since we released Airbyte.
Brett
Wow. Yeah, I would say so. That’s very incredible growth. So let’s just look at that first number. There 90,000 people there in the community. How did you achieve that? Where did that community growth come from? What are some of those lessons that you learned along the way in building.
Michel Tricot
That you have to invest massively in it, in term of your time, in term of the support that you’re giving? We bootstrap the community, really between July 2020 and December 2020, because we are doing all these calls, were keeping in touch with every single person that we are talking to and we’re keeping them in the loop about what is the progress with Airbite. Because whenever we are describing our idea behind Airbus, they were just, yes, let us know what is available because we want to try it. And when we released it, end of 2020, I would say I would almost call it a prototype. So it was barely working. We very quickly put our slack up and we started to invite people on our slack and we started to get people that were just downloading our buys, testing it, and giving us a ton of feedback.
Michel Tricot
And so the whole team at the time were not that many, were just maybe six people, seven people. We were just spending half of our time with the community and the other half of our time building Airbags. But we really put a lot of time and a lot of effort there. We also published a lot of article around what does it mean to build a company just for the community to understand when were as a company, what was our DNA? And yeah, just for them to have that visibility into what we’re doing, how we’re behaving, how we’re making decision. And I think in a way that transparency created a lot of trust between the Airby team and the community. And this is something we continuously are doing. I think recently I was on Reddit and someone was saying something about the fact that Airbyte is the one place where they go when they need learning resources about data.
Michel Tricot
Because we’re putting a lot of article out to just train data engineers to help data people to make better decision on how they structure their data infrastructure, their data stack. So always about caring for the community and just always listening to what their pain are, where they are successful, and just double down on those.
Brett
And where do you think that discipline came from for you? Is that something that you just naturally understood? Did you hire some great marketing team members? Because where did that come from? And the reason I ask is I just see a lot of companies out there who aren’t really willing to invest in marketing that’s not directly tied to their product, all they want to do is talk about their product and their capabilities and their features. They don’t really want to do this top of funnel content. That’s goal is really just to educate their potential customers. So how did you develop that discipline within the company?
Michel Tricot
Yeah, I think it probably comes from the founders and the early team, which is we’ve always had that mindset of bottom up growth. You can almost say to an extent it’s very A, B, two C approach on how we’re thinking about B two B product, which is in the world of today, a lot of the decision is driven by the user and not the buyer. And when you’re doing open source, what’s happening is that you need to win the heart of the community. And from there, you cannot just say, okay, we’re just going to do ads. That’s not how it works. You need to engage with people and you need to engage personally with people. And because were seeing open source as this is a very bottom up way of building a product, of getting adoption, it seems very natural to do it this way.
Michel Tricot
And I don’t know if I take a note as to why, but it’s just if you do bottom up, it means that the people you’re talking to are your users all the time. You’re not talking to the buyer, you’re talking to the user. You need to get them into the product. You need to love them to love the product. So then it just become how do you scale that process? How do you get them to like you?
Brett
And then is your sales process 100% PLG or is it sales assisted at some point?
Michel Tricot
PLG is great, but there is always a point, especially as spending goes up, where people need to talk to a salesperson. So, yes, I almost see PLG as something that happened before you talk to sales. It’s almost like a filter where you will have a huge category of people who never need to talk to sales, and you need to make sure you can address those, especially with the motion of Open source that we have. Like open source in a way, it’s a PLG motion, which is people download the software and they need to be successful without talking to you. So this PLG motion was already ingrained into the open source software. Now, when we did Cloud, we thought about it the same way, which is how do we maximize our footprint in the data world? And not every company needs to talk to sales. So from there, we go for PLG, we go for Sales Serve.
Michel Tricot
We make sure that the product shows its value as fast as possible. But then as people start to use the product more and more, they have a need that requires them to just spend more, to book more. Then we go for a sales assist motion.
Brett
And on the topic of open source, what would you say is the number one benefit that you’ve gained from being open source? And then what’s been the greatest challenge because you are open source?
Michel Tricot
The great thing you get from open source is that first it works extremely well with the product that we are developing, which is we need to address an infinity of data silos. And when you’re thinking about an infinity of data silos, you realize very quickly that it is impossible for one company to just address that infinity. And there will be edge cases that will not work with most people. And how do you make sure that someone who needs that edge case can still use the product? And from there, it became obvious that to scale our data connectors, we needed to actually involve the community, we needed to involve our users, and we needed to involve companies that are using earbyte. So from there, open source was almost obvious. In terms of the challenges, I would say you’re basically building two different motion, two different product in the sense that when you do open source, the type of personnel that you’re talking to might be a little bit more ahead of the curve.
Michel Tricot
They are more tinkering with open source or trying to solve the problem by themselves. And this is something that were able to leverage a lot, which is were putting stuff out, were explaining the golden path for how to use it, but people were just tinkering with it. Discovering new use case. When you start going more on the paid offering of Airbyte, the persona is changing a little bit, which is the expectation in term of reliability, in term of ease of use go up quite fast. And for us, we had to basically retrain ourselves in 2022 in how do we just think about this reliability piece, this ease of use and making it part of the product. And what’s actually interesting is that it also benefit all the open source users, which is now the product is simpler to use, faster, more reliable. So that was something that we had to read on in a way, makes.
Brett
A lot of sense. And then something also I want to ask about, I was reading about in the media how we’re able to reach unicorn status within two years of launching, which is just insane and very impressive. So can you take us back to the day that was finalized? You know, when you found out that you had built a unicorn, what did that feel like for you as a founder?
Michel Tricot
I’m going to be very brutally honest on this one, which is I’ve almost banished the world, the unicorn, from my vocabulary. The way I see it is we’ve demonstrated that the model were building especially for the problem we’re solving, was a massive disruptor for the industry. But the way I see that is not I never see fundraising or money that you get from investor as a milestone. I only see that as fuel to help you grow faster. And when I say big valuation, I don’t see that as, okay, this is what we are now. Yes, it’s great to see that it’s a good number, et cetera, but it’s more a promise for the future. And we got the capital that we actually need to accelerate Airbite and to get us to this level and to the next level. So it is great, but I think we should continue to keep our head on building the best product and delighting customers at that point.
Michel Tricot
The rest is a bit of a distraction if you think about just, okay, yeah, now we’re going to do big party with ton of cash and celebrating 1.5 billion valuation. That feels unnatural. What’s important is just building an amazing product, building an amazing team, and building an amazing community and customer experience. Yeah.
Brett
And what you’re saying there sounds very logical. I guess I just wonder, how do you build that discipline internally? Have you just always had that much discipline? Or when you started to see the company growing like this, did you have almost like an internal talk with yourself to be like, hey, Michelle, keep it together, keep calm, keep focused on the business. Don’t let this be a distraction. Was that hard for you to do, or is that just easy and totally natural for you to do?
Michel Tricot
It is very natural for both John and I to do it. And soon after we raised our Series B, we managed to bring the whole team together. We’re very distributed as a team, and we brought the whole team in San Francisco. And at some point, someone was suggesting, oh, we should do a Series B celebration. And I was, no, we’re not going to do Series B celebration, but we are going to do a team is together. After two years being in COVID celebration, this is what matter for the company. And I think people think that you should always celebrate. I mean, these are also victories, but they are not the goal. And I think it’s what is important as founders, is just making sure that people keep their eyes on the real goal, not on the decide accomplishment. Nice.
Brett
I love that. And that’s certainly very valuable advice for the founders who are listening today. Now, another question I want to ask about is your market category. So the first part of the question, what category does Gartner place you in? Is it the data integration category or what is the actual category from Gartner?
Michel Tricot
Yeah, data integration is a good one, and I actually think it’s the one we’re being put in. We’re trying to publicize more around data movement, which is it’s not just about integrating data, it’s about predefined the movement of data and the portability of data. But I think right now, the category yes, would be data integration.
Brett
And that was my next question then. Are you content with that category that you’re in? Because a lot of the founders that I speak to, especially companies that are close to your sides, what they say is they’re not happy with the category they’re in. They don’t feel that they belong in the confines of the categories that Gartner places them in. And they’re all about creating a new category and category creation. So what are your thoughts there? Do you plan on or are you currently working on creating a new category or are you going to really just move into transforming this existing category and trying to reposition it in a different way?
Michel Tricot
Yeah, I think it’s good that you put in a category in a way because you have less market education to do. Like, people know what to expect, they understand the context, they understand what value you’re creating. Now, you should not let that to be the box in which you evolve. And if you realize that there are adjacent things or a longer time vision that makes the product even stronger and that can allow you actually to create a potential new category or to transform the existing one, then you should do it. And honestly, right now it’s okay for a label as data integration, but I think data movement is a stronger way to think about it because it has more although the term is not completely fully formed or fully understood, it still has this concept of it’s more than just connecting to a data silo.
Michel Tricot
You have a lot more around that. It’s just how do you get company to exchange data in a smoother way? It’s how do you make the data possible? Like what kind of contract do you need to put on top of your data, what kind of quality, what kind of expectation you can have on top of data. And this is not so much in the data integration space, but I think for now it’s okay and we’ll see how that evolves, but makes sense.
Brett
Now, last couple of questions for you since I know we’re up on time here. What excites you most about the work you get to do every day?
Michel Tricot
I mean, there is one, we have a Nostalg metric at Airbyte that it’s volume of data that we’re moving basically. And we have it separated between cloud and within open source. And the reason we picked that North Star metric is because it’s not a vanity metric. It is synonymous with the value we provide to our customers and with our community. And this is something I really enjoy doing is finding these things that can rally the team behind something that seems like a very simple goal. There is complexity behind but just keeping the focus of the team on this. One thing that we’re doing, which is this is the value we are bringing to our customers and let’s focus on it. The rest is vanity, the rest is a. Distraction. If we get the volume really high, then, yes, revenue is here. Yes, the community is growing, and, yes, we’re building a valuable product.
Michel Tricot
So just finding this little pocket of wisdom and just alignment for the cinema is something I really enjoy.
Brett
And I know you touched on it there when I was asking about the valuation status that you achieved and that there’s a big goal that you’re working towards that you don’t want to be distracted from. So what is that big goal? Let’s talk about the three year vision. What’s that vision that you’re racing towards and what are you building for?
Michel Tricot
And that will go back to why the community is so important, is when you go and ask a software engineer, oh, you should start that project, the first thing they’re going to be thinking about is, okay, I need to start my project on GitHub, or I need to start my project on GitHub. So you have created a standard for what does it mean to do software engineering. We operate more in the data space, but our goal in the next three years is establishing Airbyte as the moment you buy your warehouse, the moment you need to move data from one place to the next, it’s clear in everybody’s mind that the default solution is a byte. And, yes, there will be also vendors here and there, but the default solution is a bite. When you were looking ten years ago as well, people wanted to do big data.
Michel Tricot
The solution was, okay, we need to use Hadoop, and then it becomes, oh, yes, we need to use Spark. It’s just how it’s about really becoming that standout so that it becomes weird to choose another technology. Nice.
Brett
I love that. All right, unfortunately, we are going to have to wrap here before we do wrap up, though. If people want to follow along with your journey as you continue to build and execute, where’s the best place for them to go?
Michel Tricot
Yeah, so two places. First one is our slack. We have a public slack. Anyone can join it. We have links on our website, slack Airbyte.com, and everybody from the company is on the slack. So you can talk to anyone at Airbyte. So I would say that’s probably the best place to start.
Brett
Amazing. Well, thank you again so much for taking the time to chat, talk about what you’re building and share this vision. This is all super exciting and look forward to having you back on again in the future to talk about everything that’s happened.
Michel Tricot
Wonderful. Thank you, Brett. All right, thank you. Keep in touch.
Twitter
facebook
linkedin