Transcript
[00:00:00] Victoria Melnikova: Hi everyone. Welcome to Dev Propulsion Labs, the podcast about the business of developer tools. I’m your host, Victoria Melnikova. I’m the head of new business at Evil Martians, and today I am ecstatic. I don’t think many people get a chance to sit with you like this in the room, so I’m very happy to introduce today’s guest.
Please meet Paul Copplestone, CEO, and founder at Supabase.
[00:00:28] Paul Copplestone: Thanks for having me.
[00:00:29] Victoria Melnikova: How are you today, Paul?
[00:00:31] Paul Copplestone: Well, if I had known that this was so professional, I probably would’ve had a shave and want something nicer. But yeah, I’m doing great. It’s nice to be here and cool that we can do this in person.
[00:00:43] Victoria Melnikova: It doesn’t get nicer than that, just so we’re clear.
Nice. You know, I put my best outfit today and Paul did too. And the reason why I’m wearing this is because. We went to Supabase Select, which happened a couple of weeks ago, and I had a true kind of like fan girl [00:01:00] moment. I really did. And I don’t think you get to hear that often from, from women because I mean, most of the builders I guess on Postgres are probably men.
But still I had a kind of like a pop. Pop culture moment at select because you and, and you kind of held the stage the whole day and you had some amazing speakers, people that are probably inspiring to you as well. You know, because they’re leading generations basically of new builders and they kind of set a certain level of excellence in tuning and products that they build.
And I’m sure it was a conscious choice. Who you bring to the conference as speakers. So let’s talk about Supabase Select. Cool. When did you come up with this idea and how did you choose the speakers?
[00:01:52] Paul Copplestone: I think everyone eventually aspires to have their own user conference and for us, we have discussed it for [00:02:00] many years and we kind of internally have a no conference policy.
Like as in we don’t attend many conferences. We just put our heads down and do work. Then this year, for whatever reason, it wasn’t me who chose it, the momentum built and mm-hmm Someone decided that now was the year, and I guess there was enough consensus built that we should do this. And then the idea, really, it was a few people internally who have just done a great job.
What had happened though, I, I guess for my part in it was I had been to the Stripe and the Figma conference. Yes. And these ones were in the Moscone Center, like, um, at the same week and then both opposite each other. So I saw how cool these conferences were. And I especially loved Figma because, you know, I was lining up and there were people turning up from all over the world, and they’re flying in and from Australia or from wherever, and they’re all dressed, you know.[00:03:00]
Their audiences, designers. Yes. And when you go outside, it felt like a festival, but it could have been something like that, a DJ playing. And I just thought, ah, that’s cool. Like it’s a place for designers to get together and hang out. And I don’t know how much they cared about hearing what everything was being launched, but why did people want to fly around the world to be part of it?
And it’s because their community is so, so big. And it’s kind of a way for them to give back to that community. This is something we have, fortunately, a really nice community with lots of builders. And so a lot of the things that we wanted to do was capture that same vibe in our own conference.
And in particular, we love the builder audience, like what Figma did for designers. We wanted to do for, for builders, developers, and builders. ‘cause I, I think the term is developer is becoming looser and looser. And so, yeah, for the speaker list, largely for [00:04:00] me, I reached out to a few of the builders that I wanted to chat to.
And ask questions about how they built their developer audience. So I was very lucky to speak to Patrick and Dylan, Patrick from Stripe, Dylan from Figma, about their early days. Guillermo as well, Al and, and spoke to a few people like James from Firebase, which was also very cool. Just to share with our own community how closely we are collaborated.
In the early days,
[00:04:27] Victoria Melnikova: it felt very handpicked. You know, it’s like, it’s very rare to be at a conference where you have such a stellar lineup and you want to hear every single talk. ‘cause it’s just like, and you are asking questions selves. So that added like a very personal touch because. I could imagine that you could have that kind of conversation with them somewhere else, you know, not at the conference.
And it was cool for me to kind of like be part of that experience. And it did feel very like niche and, I don’t know, just like really nice. [00:05:00] It was one of the best conference experiences I’ve ever had and I’ve been to many. Ah, thanks. So, congratulations. Thank you. And obviously you made a few important announcements during the conference.
When, if we talk about business side of things, it’s interesting because you announced multi grass, which is a big deal, and it means that you are pursuing bigger clients, right? Like Enterprise. How do you see Supabase now with announced series E? Like where it’s going? What? What do you envision for Supabase in a couple of years, in 10 years?
[00:05:38] Paul Copplestone: We’re very focused on data problems and so we, we announced our series E, we did this round and one of the things that we wanted to do as well was bake in a community round. Actually, I’ve wanted to do it for many years, but each time we raise it, then investors and lawyers say, ah, it’s a [00:06:00] lot of work. It’s a lot of work.
Which it is actually Will and, and you have some regulatory things that you have to, Meat. So probably like we’ll raise only a million dollars to, from the community to fly under the radar and it’ll probably cost us that much ‘cause we were to do a bunch of audits. And things like that. So it’s less about raising the cash from them, but allowing them the community to invest if they can.
I would’ve loved to do it sooner so they get a better price. But to be honest, we we’re such a small org and nimble org that we haven’t had the resources. So now, yeah, we’re doing what we can and hopefully it becomes a more regular thing. Let’s see when we’ll do our first one and see. See where we land.
And then yeah, using the money. We’re still very much in this stage of the company where we want to build out a lot of this data platform and the infrastructure. So like you said, we’ve got, you know, a focus on the enterprise, but to be honest, also a big focus on a lot of our scale up [00:07:00] customers who.
Get started with superb base. This is our tagline, build in we and scale to millions. And we do have these customers that scale up, you know, within a week they scale to a million users or, or, or beyond. And they eventually hit the kind of upper limits of Postgres. Which it’s very scalable, Postgres, but when you start reaching its upper limits, you’ve gotta do things to make it work.
And you’ve gotta have a lot of handholding. For us, we are a very PLG company. Product like growth, and we see a lot of that handholding exercise should just be product and yeah, as much as we can, we’re trying to build around the maybe limitations or, or the things that Postgres is lacking for now, and also contribute back as much as we can to the Postgres ecosystem so that others can, can use Postgres confidently even at at scale.
[00:07:56] Victoria Melnikova: So if you think about Supabases as a product, there [00:08:00] is Postgres, there’s open source, and Supabases provides this service for people, right, that they can use this database and the power of Postgres without any additional kind of like brain work. If you think about the recent growth that happened due to this increase of pipe coating, this new kind of trend.
Do you feel like the product offering is changing as more of those people on board to superbee or is just part of the spectrum of people that are using and eventually they’ll become coders or like more serious builders?
[00:08:37] Paul Copplestone: We haven’t specifically designed for like a vibe coding profile. One of the ways we thought about which profile to build for in the past was say if you take a spectrum of front end developer.
To hardcore, Postgres engineer probably you wanna be a little bit tw like 25% down the spectrum. And now the [00:09:00] end of the spectrum has changed. So it’s shifted out not just like a front end engineer who doesn’t know a database. Now you’ve got a person who doesn’t even know what front end and backend might mean.
You know? At the same time, what we were having to do in the past was, let’s say we wanted to expose a lot of complexity in, in Postgres. Um, maybe around role level security policies or something similar or how to write triggers. And what we were thinking of doing a lot was building a UI to kind of step people through maybe a wizard based approach or Yeah.
Like surfacing a lot of these unknown things through. A nice user interface. And then we would also expose the developer to some SQL. So that they were progressively learning SQL along the way. So a lot of people learn SQL using superb base, which I think is good. It’s kind of like a superpower.
Now the nice thing, even as the [00:10:00] profile of developer has changed, SOTU has the way which we expose the tooling. So people are becoming very accustomed to a chat based interface. Which meant that we could pull out a lot of these UI elements and now we just have, say, a contextual chat and SQL and these two modalities fit those two sides of the spectrum quite nicely.
And again, they can get gradually exposed to SQL and you can say, Hey, I need to build a database. Or a library of my Pokemon cards, or whatever it is that they want to build, and it can scaffold out this whole database schema for them. Show them the schema before they run it, and maybe they just hit run, but at the very least they’ve seen it.
They know that something’s happening. It’s not magic. And this is nice. They get as much as we can. We’re trying to gradually expose people into the complexity that, that they have access to.
[00:10:55] Victoria Melnikova: With my previous guest, we talked about Steve Jobs and how he [00:11:00] talks about everything around us, that it’s built by people and you can actually impact everything that’s around you.
You can build your own and you can create a new experience. You know, you can create new journeys for people living. As a founder, co-founder of Superbs, do you feel like you’re impacting like thousands of lives? Like do you experience that on a daily basis or it’s kind of like, you know, we’re just doing our work.
No biggie.
[00:11:31] Paul Copplestone: I mean, there’s some rare moments that I get outta my house and maybe I meet a friend who I met one when I was in Switzerland a couple of months ago, and he said, I, I hadn’t seen him in a few years and I. I, I said, oh, yeah, I’m back into building another startup now. He’s from the startup world.
And he said, oh, what is it that you’re building? And I said, superb base. He’s not a techie. And he, he said, oh, I know. Supabase. So this is very cool. Like Suddenly the data, [00:12:00] the, the, my startup. Which usually is too technical for non-technical audiences is a bit more of a, I guess, household name amongst people who build.
And that’s nice. I like to see that it’s solving it. A lot of what I’m motivated by is just solving the problems themselves. So I’m quite often out in the community online just seeing what people have problems to solve. So as much as it is nice to have the recognition of. Say someone’s using it. I actually get a lot more recognition seeing that they’ve got a problem and we can solve this problem.
Or not, we do the best we can, and then actually kind of overcoming that problem, which could have been a challenge for them. So this feedback loop between hearing something, developing And succeeding to solve their problem, that that’s where I get a lot of my motivation.
[00:12:53] Victoria Melnikova: What does your job look like day to day?
[00:12:59] Paul Copplestone: Yeah, it’s a good [00:13:00] question. Um, it changes now day to day where I used to be quite locked in and wanting to code all the time. It’s now more of a liability. If I’m coding because I can’t do any sustained work. It can be anything from yesterday, US East went down. It takes down half the internet. And for this, maybe not everyone understands what that means, or people point out different things online and I have to, you know, correct the record on on some things.
So a lot of community management, likewise doing these things, if I’m traveling, if I’m outside my office or just catching up with the, the team. So a lot of our, uh, this year we’ll double our team, which is. We could do a lot more, but that’s about the healthiest we can do it. And if we do less, it’s also going to break things.
So, you know, when you double in size, yeah, a lot of things [00:14:00] change. You have to reinvent a lot of things. Quite rapidly, and that change is difficult for humans. A lot of the processes that you started even six months ago might not be relevant anymore or useful, and so you need to change them. And some of those things could be useful to a small portion of people, but not to everyone.
Now it’s really about trying to figure out how to navigate that within the org because we need to grow. How? How support tickets have grown. Yeah, five or 10 x and we can only grow the support team. Two x. So, yeah, it’s, it’s finding a balance and it’s a difficult stage of any startup. I think any hyper, like anything that grows this fast has this period and doing it as gracefully as we can is one of the, one of the things that I think a lot about,
[00:14:48] Victoria Melnikova: one of the things that you look for in people when you hire is low ego, and you also have a fully distributed remote team.
Does that complicate things as far as the. [00:15:00] Exponential growth goals?
[00:15:03] Paul Copplestone: A bit of both, I think. I feel sorry for, for new joiners, it’s very difficult to learn and know your colleagues when you’re in a fully distributed team. And just for those listening, like we don’t have any offices, we’ve barely got meetings.
Like maybe one meeting per team per week is kind of the, the, the guideline. So you could, if you wanted, you can turn up to any meeting, but you could. Only do 30 minutes FaceTime with your team per week. And so it can be difficult to integrate into superb base. At the same time, there are huge benefits.
People can focus. There’s no travel. You can work anytime you want and you just focus on results. For culture especially where there’s friction, there’s benefits to everything being online. So for example. It’s very hard with this many people. We’re in, we’ve [00:16:00] got people in 40 different countries now, I think.
And. Because we don’t care where we hire from. We just, if the person’s the right for the job, we hire, hire them. But 40 different countries means 40 different communication styles. And so of course getting people to work together with different styles Even with the same style can be difficult.
The nice thing about it being online is you can quite easily course correct on people’s communication styles because everything’s written down. If someone says something, which can be perceived to be bad, I don’t have to, they don’t have to come to me and say, oh, they said this and they said it this way, and I have to go to the person awkwardly and say, Hey, sounds like you said this, and they misinterpret it.
I can literally link them to the slack thread and say, Hey, just so you know, if you say it like this. It can be better. So there’s just pros and cons. The most difficult, of course, is alignment. Across [00:17:00] organizations. Like we can’t get in a room and talk about difficult problems. And so that’s the challenge.
And the way we solve that largely is trying to hire very senior people who look after their own roadmaps, do a lot of bottom up, and then very light on the processes. So making sure that. We can adopt processes across the entire org. With a few changes, a few tools as well, so that we are not scrapping over which tools to use and which teams just trying to consolidate the basic stuff.
[00:17:34] Victoria Melnikova: So let’s talk about you un and it’s kind of crazy, right? So you are from the opposites of the world. You met in Singapore. The two of you. He’s from England, you’re from New Zealand. Same language, but everything else is very different.
[00:17:50] Paul Copplestone: My, my wife would disagree on that. She’s from France and for her we’re basically the same culture.
[00:17:58] Victoria Melnikova: You do seem very [00:18:00] an organic fit for each other. How did you find each other? How did that happen?
[00:18:05] Paul Copplestone: Yeah, it was really cool. We’ve actually, a few people in Supabase came to Singapore and we all lived together. The way that it happened was for me, I did my first startup in Southeast Asia and it was across.
Three countries. Singapore being one of them. And then after I wrapped up I thought, ah, what am I going to do next? And I was thinking of coming to the US but my friend was recruiting for this thing called Entrepreneur First. And this is an accelerator where you get in a room, you meet a hundred people, and then you start an idea.
But it’s all, I don’t know if this is still the case, but at the time it was very deep tech. So things like Space Tech. Quantum computing. I was working on virtual reality, so a very interesting group of people and at the same time and moved from, he got into entrepreneur [00:19:00] first and they said, well, you can do it in London or you can do it in Singapore.
And he said, wow, I want to do it in Singapore. That sounds awesome. So we mo both moved there at the same time and did the course together. He ended up doing the company through Entrepreneur first. Because I didn’t think the idea was going to happen fast enough. The timing wasn’t good, and we ended up moving in together in a hacker house where there, I think there were seven of us and maybe we’re building six different companies.
And then I built my second startup. And in the process I was kind of building what is now the tech behind superb base. And when I wanted to do it, I thought, all right, I want to go to yc. I wanna build this database startup. It’s something that I’ve wanted to do for literally over a decade because I, I really like Ant and he’s such a good human.
One of my key things was that I wanted. To have someone that we would spend, you know, a decade working [00:20:00] on this with. So yeah, I asked him, and fortunately he said yes. And yeah, here we are. It’s been, yeah, five and a half years now working together. So
[00:20:09] Victoria Melnikova: seven, seven years left. For the, for the, for the decade complete.
Like even go in further back growing up in New Zealand, use a startup ecosystem in New Zealand. Like active, like how do you learn? Oh no, how do
[00:20:22] Paul Copplestone: you
[00:20:23] Victoria Melnikova: learn
[00:20:23] Paul Copplestone: about that stuff? I didn’t even know what a startup was. I, I was going to present this at the Supabase Select, but I found an old email, I think it’s from 2008 or 2009, where I had written to.
I wanted to build this database startup, and it’s like I’m gonna build it for developers and you can both design and develop the database at the same time. It’ll be very visual. Basically describing superb base what it is now. And I had sent this email to just someone who I knew was a very rich tech entrepreneur, like kind of pitching him the idea and he quickly shot me down.[00:21:00]
I was said like, that’s, no, I don’t wanna fund it. But at that stage I didn’t know I, I knew entrepreneurship. Yeah, because my father’s an entrepreneur, but I didn’t know of the startup scene and that you could raise money and venture capital, any of these things I just wanted to build. So it wasn’t till many years later that I tried to get into the scene and then that’s when I moved, like I moved out of New Zealand and I met someone eventually who.
For my first startup was the CEO, and he needed someone to do the tech, so I moved to Kuala Lumpur to do that.
[00:21:35] Victoria Melnikova: I also wanted to talk about marketing a bit because, so I mean Supabase is famous for, for tech marketing. Some people would even say that you originated the culture of Launch weeks, and I don’t know if it’s true, actually.
Probably yes. Right? Like you were probably one of the first one, the first ones to do this and to give it a name. Launch week as a concept [00:22:00] is something that you started intentionally, or it’s just like you found out that it works for Supabase. You made kind of like a concept out of it and gave it to the world.
[00:22:12] Paul Copplestone: Yeah, and that that website, by the way, is maintained by one of our communities and it’s doing a really great job. We thought about using the name Launch Week Dev to kind of for ourselves, but then I realized, oh, it’s so much nicer if, if he wants to use it and he does a newsletter and, you know, so if you wanna follow along with all the other developer tool launches is worth checking out.
The way it happened was that we, we kind of got this culture of kaizen, um, yeah, it was from the Toyota production system where you, Just to continuous improvement. The idea is from the CEO to the assembly line worker in In Toyota production system. So incremental changes on the time. And so what we did was we did a couple of launches during yc.
And then we saw how [00:23:00] successful each launch was for the kind of slope of the curve, where each time we did it, we grew a little bit faster when we got out of yc. Demo day was such a good forcing function for us. We had this thing in front of us and we wanted to ship, so we said, all right, let’s just emulate our own.
Like, let’s just set a date in the future and we’ll try ship something. Then, so we just ship, ship, ship, and then we, what we did was we went from alpha to beta. On that date, nothing really changed except for that tagline. Alpha de beta again. The slope of the line changed just ‘cause we kind of announced, we put down now developer hats and we put, picked up our marketing hats.
So then we thought, all right, this is working. What do we, how do we do it a little bit better, just in a kaizen manner? And we said, all right, let’s try to ship one thing every day for a week. Instead of one launch, let’s do five launches. And we came up with eight things and you know, even if we end, end up only shipping three things, then it’s [00:24:00] better than one thing.
So in the end, we got those five things out and. We called it a launch week because we had five things, and from there we just continued doing them because we saw again that it was very successful.
[00:24:14] Victoria Melnikova: Does it still work for super bees?
[00:24:16] Paul Copplestone: It does. If nothing, uh, so nowadays the slope of the line kind of takes care of itself, and there’s many other things that affect it.
The main thing that I think why we do it internally now is because it’s still a good forcing function. We don’t put timelines on our team. We have this concept of. Fixed timeline, variable scope. So we like, if people wanna ship something, it’s less about what they ship and when they ship what, like, we’ve got this launch week coming up, what do you wanna ship?
And they’ll try do it. But if it misses it, you shouldn’t just ship because it’s half baked. And yeah, we said that date, all we care about is like you’ve got a bit of momentum or pressure to organize things. And also. [00:25:00] If you leave developers unchecked, they probably won’t just keep shipping and never announce what, what they’ve done.
So it’s nice for them to come up and take a breath of fresh air and you know, chat to the community. And we even do things that aren’t that successful in launch week at like at Twitter spaces maybe only has like. Say 150 people turn up or something like that. Or maybe a hundred people turn up, which by the absolute size of our company, you would think maybe would get thousands of people turning up.
That’s not what it’s about for us. We just, if we see it grow each time and the developers are chatting to the community and they’re also practicing how to, you know, be a more product minded engineer. All of these things are useful. Not for the outward facing thing, but just to build a more holistic organization.
[00:25:50] Victoria Melnikova: I remember the last time we spoke in like a few years ago, I told you that I’m a huge fan of memes. Like I [00:26:00] just, I die for it. It’s my, it’s my thing. And you mentioned that it’s aunt’s responsibility, right? It’s his domain. He is the boss. And he did that on stage too? So I’m like, okay, now I believe it.
Now I believe it. Is it still his domain?
[00:26:16] Paul Copplestone: Yeah, they do. They do workshops together. I knew meme workshops, but, but I mean, it’s very hard to hire for, for that type of, yeah, I could imagine.
[00:26:28] Victoria Melnikova: AI is not, no, AI is
[00:26:30] Paul Copplestone: terrible at memes. I mean, you can use it to generate some content, but not the, you know, memes are Yeah.
They’re really one of these ineffable type of things. You don’t get why you, you get it, but it’s like standup comedy. You can practice a little
[00:26:44] Victoria Melnikova: bit. So you do workshops, internal workshops and
[00:26:48] Paul Copplestone: runs them? Yeah, yeah, Well, them and the marketing team, I assume we’ve got a lot of good meme people and we do have a very, our, our Slack threads are very organized.
So we’ve, they’re [00:27:00] all name spaced and things, so like external or like Admin or whatever it is. Teams. And there’s only a few that don’t have organized names, and that’s our water cooler. And our, our meme channel, these are the two where you can kind of, they’re a bit more freeform.
[00:27:20] Victoria Melnikova: So you, and you kind of split responsibilities between each other.
And you mentioned that at a previous startup you were A CTO, and now you’re A CEO, which is a more, I guess, business side. How do you split your responsibilities now? Like do you feel like you compliment each other with your skill sets?
[00:27:41] Paul Copplestone: We’re very complimentary and we have a lot of overlapping ones where, you know, we both have opinions on stuff, but maybe the other one can, can make decisions.
And we’ve got other people as well that I would see as co-founder, like who also have complimentary and their opinions are very important to us. And [00:28:00] even like the early employees, we’ve just got so many of these because we hire a lot of ex founders whose opinions like I value above my own. And a lot of the things that they’re doing day to day.
I guess we have a very flat org, and between Anne and myself, maybe our titles don’t really reflect our day-to-day job, so Maybe he looks after a lot of the marketing, of course, and also a lot of the hiring. And yeah, a lot of the culture as well. And that’s where maybe we overlap. Culture is one that’s important for many people.
Just especially if you’re a co-founder, even the way you behave. Is going to seep down whether you like it or not as part of your culture. For me, working with people internally, if and when I can, and if things are, there’s a lot of friction and then I am particularly interested in product and strategy, making sure that we’re building for the next five years and, and just trying to skate where the puck is going.
On the product side, [00:29:00] working with customers who have had friction. Around Postgres. And then, yeah, a lot of the community stuff, especially when things are, I guess more in crisis mode than me as A CEO, I should be focused on, on taking on that responsibility as well. And then fundraising, of course, the mm-hmm.
Some of these business things and hiring leadership, uh. A bit of both on, on the side.
[00:29:24] Victoria Melnikova: So you kind of take pride in hiring ex founders and I understand why. You know, and you talked about it at, at Supabase select also hiring people that can pick a domain and be in charge of it and kind of like take it almost like high agency unit within the org.
How does that combine with a low ego like founder and ego seems to be like, you know.
[00:29:49] Paul Copplestone: Yeah, well, I guess controversially, they probably should not, should not conflict, but I can see why like, it might be seen, first of all, you know, if a founder [00:30:00] wants to join Supabase, you know, not as a co-founder, then probab.
Like, if they don’t, it’s a self-selecting thing. I, if you see what I mean? Like You want to no longer be a co-founder. Yes. You don’t wanna be in the limelight. I think a lot of the people who joined who were founders realize that it’s not that glamorous. Like, to be honest, I wouldn’t mind after Supabase is done going, working for someone else and spending 40 hours a week just developing product stuff,
[00:30:28] Victoria Melnikova: would you join Supabasees?
I would love to join Supabase.
[00:30:31] Paul Copplestone: And just do that. It’s not as glamorous as it seems to, uh, and you, you are the face of things, but that matters little actually when, There’s a lot of purpose to be found in just solving a problem and doing it right. The other thing is that, you know, a lot of the successful, super successful founders no longer want to work, so it’s, it’s [00:31:00] less that type of fun mentality and a lot more who might have.
Know, I’ve been grinding for the past six years and couldn’t find product market fit. And there’s something special about that type of profile. Joining because by necessity, their ego’s been beaten out of them. Anyone who’s done this grind me, I’ve had to do this. And it’s, it’s hard and it’s probably one of the hardest things in the world to, to find product, market fit and grind on it for, for many years.
And you can have the best product, the best team, and sometimes just the timing and your luck does not go your way. This time Supabase had a lot of luck and you know, we, we have a phenomenal team and we have a, a great product, but you know, the timing was just right. And so there’s nothing to say that like someone else couldn’t have captured this opportunity for this time we did.
It’s nice that we can share that with other ex founders who come in realize, you know, they appreciate the up into the right charts. Much better than [00:32:00] someone who’s coming from a FANG startup and they kind of. They’re used to, you know, billions of whatever every single week. Whereas the person who’s been grinding for a long time come in and they, they appreciate the opportunity so much more because they’ve realized how lucky it is to be in this position.
[00:32:19] Victoria Melnikova: It’s interesting you bring me to this point once again with Vibe Code and, because it’s kind of like a phenomenon that we see with Bolt, you know, and we’ve been working with Stack Lids for, I don’t know, a few years, like four years. And I remember, I remember this moment when Bold started taking off and we even had like a shared Slack channel.
And I’m like, is, is this Supabase button coming? And you’re like, yeah, we’re working on it. And I’m like, this is super exciting. But the reason why those companies chose Supabase to be the number one database for their products is kind of a mystery. Like was it your personal [00:33:00] network reaching out to them or was it just.
The state of superb base for the moment when this enterprise happened, like what was the reason why superb base became that number one default option?
[00:33:11] Paul Copplestone: A lot of superb base’s success is just focused on building a good platform, and one of the ways that we build a good platform is we try to make it extremely integratable our API.
Is consumed by our front end, our front end. Then you know, ex, the API team have to design it in a way that can be accessed by integrations. We also settle on standards like Postgres as a standard. We were a complete product. We knew that the AI we had been building for AI for for many. Years at that stage mm-hmm.
Where like, we had been testing things out and, and consuming it and, and, and focused on what would the next wave be. So it’s probably a series of things. And I know for example, they actually tried other platforms and didn’t [00:34:00] have as much success at getting the LMS to, to build with it. So, you know, even just the fact that we had a burgeoning community that had discussed what works and what doesn’t work online in Reddit on GitHub.
We, you know, even put our troubleshooting guides on GitHub, so they’re consumable lots and lots of small things that eventually led to a moment where it was, I guess, just designed for the right moment, the right point in time. So, yeah, again, like luck is a, a funnel that you try to widen as as much as you can.
And in this case, we were lucky that they chose us. I feel very fortunate that they settled with us and it’s been a great. Boon to superb base to have them consuming us. And, and lots and lots of people throughout this year building on top of superb base. But yeah, full credit to the team for, for designing it in this way where, where it can be consumed and integrated with, uh, even if you are [00:35:00] building a platform on top of us.
[00:35:02] Victoria Melnikova: There is this opinion that the vibe code in bubble is gonna burst. You know, that is just like this trend, it’s hype, whatever. But in, you know, in a couple of months, in a year, it’s gonna kind of like go down. Do you feel like that’s true or do you think that AI in general is making, building more accessible to people who don’t necessarily have coding skills?
[00:35:24] Paul Copplestone: Of course. Like when you see the debates online, they lack a lot of nuance and they just think, oh, bubble hand. That’s it. Vibe coding is here to stay. That’s one thing that I can tell from From, yeah, my interactions and what we see, the form in which it might happen in the next few years, and maybe the pricing mechanisms around it and like how it interacts with, say, existing code bases versus maybe as a prototyping.
All of these things are where I think the discussion is much more fruitful. I think, you know, vibe, coding. At this [00:36:00] stage, I’ve talked to product offices at very large companies and they love these tools and they, they’re able to do things that they previously couldn’t, and they get more velocity from their entire org because they can build things and show off things and make videos.
Now, will it be able to attach to a huge code base? I think then we’re going to see a lot of the, these two worlds merge a little bit where, Where the vibe coding tools need to exist. Sit on top of existing enterprise workflows. And likewise, uh, these workflows will probably want to pull in a lot of more of this, say AI based prototyping and developing and maybe even developing features or testing or whatever.
So I think the form, like this idea that is one thing here, and will this one drop away and like will go back to being software developers. And that’s the only way the world will operate in the future of the bumble burst is a bit naive. So [00:37:00] I’ve got some more nuanced theories around how am I looking in a few years and which is basically trying to mash these two worlds together.
And so yeah, whether the bubble bursts or there’s a correction around some of the economics And the companies is a, is a valid conversation around maybe the economics of it and who’s going to use it. But I think that the trend is here to say stay, and there will be winners in this space for sure.
[00:37:29] Victoria Melnikova: Does your vision of Supabase, like let’s say in 2020 when you started Supabases, could you imagine this, what’s happening today for Supabase? Or was it always your plan to be like that big, to have millions of users and you were just like looking for the right opportunity?
[00:37:48] Paul Copplestone: Our thesis at the start was.
Pretty simple. We wanted to build a database company, and if you look at what almost every database company has done in the past, say, 20 [00:38:00] years, generally they build a new database which is bigger and better, and then they try to migrate people to it and. It’s not a bad idea. Building a new database, which is bigger than and better than some other incumbent one that that actually is, it works in a lot of cases.
What doesn’t work is migrating people to it. Yeah, because it’s very hard sticky. It’s, it’s very sticky and. It’s the only company that I’ve seen do very well in this case, well, maybe two, were both Firebase and MongoDB because they focused on a new trend, something that changed. And they also were very good, especially at the start with MongoDB, at getting in on greenfield workloads.
Then it turns out, if you can do this over the course of a generation, then you become the generation’s database and you will, you earn the right to have the large databases. And so we knew that if we were to build the largest database company in the world, you know, you want to take [00:39:00] on the Giants.
It’s not to go head to head with them at the start, we’re going to Be in this for. Building for 30 years, you know? And we need to build out some of the incredibly hard systems that we’re starting to develop now, but we’ve earned that, right? By focusing on these new workloads. We’ve launched 10 million databases.
We, we have 5 million developers. Like, these are the things that we have now earned the right to, to, like, it’s ours to lose at this point.
[00:39:30] Victoria Melnikova: I have so many questions and I’m, I’m kind of torn because I want to ask personal questions, you know? Like, for example, is there a work life balance for you? Like is, do you have a life, you know?
Because running a startup, especially a successful one, is a big, big job. And I also want to kind of get this opportunity to, to ask you for advice for aspiring founders, you know? Because. I don’t want them to feel like [00:40:00] what they’re working for is stripping away the balance from, from life, you know?
So I’ll ask both questions then. Is there a work life balance in your life?
[00:40:10] Paul Copplestone: Work life balance is slightly harder, I think. Building a startup is a lot of work. At some point you can add more leadership, which diminishes the amount of time you have to spend. In the team, but not necessarily on the stress.
There are, there are always different problems and so yeah, it’s definitely difficult, but it becomes easier as you hire good people. Mm. That’s, that’s, that’s one thing that I think now we’re in the stage where we’ve just got very good people who we really trust. The comradery makes it much easier to do, and having a good founder, a co-founder like Ant is incredibly important for, for the staying power of.
To do this for the next 10 years, you know? So [00:41:00] I’ll be honest, no, there’s not, not a lot of work life balance. And maybe you have to choose one or two things that you can do outside of work and yeah, commit yourself to those things. So you have do have something outside of work, but, but yeah, you’ve gotta work.
Hard and smart. I will say that I probably worked harder on my first startup than I did on this one. Interesting. And, and it didn’t succeed. So it, it’s, it’s not necessarily that you can just, you work hard and suddenly that’s the thing that makes you successful. There’s many things that you have to do.
Right. Working hard is one of them, but it’s not the major thing.
[00:41:38] Victoria Melnikova: So if you were to talk to somebody who is about to build Let’s say, You yourself, I don’t know when you met and at that time. What would be this opportunity that you feel like is a good opportunity to jump on? What would you advise to pay attention to
[00:41:58] Paul Copplestone: today?
[00:41:58] Victoria Melnikova: Yeah,
[00:41:59] Paul Copplestone: that is an [00:42:00] incredibly difficult question. I don’t know if I, if I wasn’t doing Supabase, I dunno what I would do today. Quite honestly, you need to action creates information, so I’d probably start something. And I’ve got two minds here. Like often I have conflicting ideas, which you just live with.
So the one conflicting idea is that really, if you raise a lot of money on an idea that is not yet proven, you tend to behave like a startup without being a startup. And this is something that I did learn in my previous startups. You have to behave at a stage appropriate manner. At the same time, I think it’s stupid to sit out a bubble we never do.
If money is available and you can can capture it, take it, but you should still behave appropriately for your stage. If you want to do some AI startup, take it, but then don’t go silly on the [00:43:00] first idea. Really, you’ve gotta hone in and find that product market fit and build towards that before trying to do the blitz scale strategy.
[00:43:08] Victoria Melnikova: So when you found product market fit at Supabase, was it right away? Was it that Hacker News, uh, post? Yeah,
[00:43:16] Paul Copplestone: product Market fit people feel is like a one thing. I feel like there’s many stages of product market fit. So in the first one we repositioned from real time Postgres to open source fire based alternative, and that was.
Three months in or something, and then we could see, you know, there was strong, strong interest, but not necessarily a lot of signups because the platform wasn’t ready. And then we went, we launched auth, and then we could see some growth that was growing organically at peon G. And then, you know, I feel like we did things throughout the journey, which again, changed the slope of the curve.
Really trying to find what are the things that we need to add to game [00:44:00] product market fit. And it’s not always intuitive what those things might be. People might be asking for things and you think, oh, I need to, yeah, add this and that will solve everything. And you add it and it might not do anything.
Or vice versa. And sometimes, I was talking to, yeah, an open source maintainer today about this one. Like sometimes we’ve added things. We might not see much use usage of that thing. But the fact that it’s there made people feel like we’re a more complete product. So it’s a very hard thing to know, like what you need to do to get product market fit.
But a lot of it’s just listening to your customers knowing to tease out what is the noise and what is actually the signal and not believing everything that, that they say. But eventually you just feel that things are. You go to sleep in the at night and you wake up and things are bigger in the morning.
[00:44:54] Victoria Melnikova: So it’s actually like, you can’t second guess it when it’s product market fit. [00:45:00] You understand?
[00:45:01] Paul Copplestone: I, I feel for us, like, you know, maybe at each stage I would say it’s product market fit, but every six months it feels more confirmed, you know? So, yeah, so, so there’s definitely a stage where it feels like it’s de-risked.
But the feeling of product market fit definitely solidifies more and more as you, as you continue and you continue to get more feedback and iterate on what the customers really want.
[00:45:29] Victoria Melnikova: So we arrived at my final question, which I call a warm pauses question, and I think I even asked you last time, I’m not sure.
So what makes you feel great about what you are doing today?
[00:45:41] Paul Copplestone: I guess people wouldn’t say this, but like, or probably shouldn’t say this, but I’m quite proud of the com of the team we’ve built and the people that they are. And I think it’s hard to build a team where, you know, you have a lot of good humans and that, [00:46:00] you know, are really focused on building and pride isn’t always seen as a, as a great thing.
But if there’s one thing and as well, like really. A lot of this has to do with ANT as well, but we’ve just focused on keeping good people, and I like that, that it makes me want to turn up to work every day and solve these problems for customers sometimes who are a little bit unreasonable. But you know, it makes it so much easier that we’ve got a group of people that are all in it together and we’re kind of working on these hard problems together.
So I think that’s, that’s my warm, fuzzy answer.
[00:46:37] Victoria Melnikova: Lastly, I want to provide stage for you to invite people to try superb base. I mean, I’m sure a lot of our listeners know how to do that, but if they didn’t, how could they find Supabase?
[00:46:49] Paul Copplestone: Yeah, if you wanna attach a database or a backend do anything, then you can just go to superb base.com, or if you just wanna follow along and [00:47:00] self-host it, even.
We’re happy with that or get involved in the open source community. We’re always looking for more contributors and we’ve got a team that can help contributors get started with getting involved in the ecosystem. So there’s many ways to get involved and yeah, I think if you’re a builder and you’ve got any feedback on, on our product after you start using it, the easiest way to get in touch with me is just DM me on on Twitter and.
I love hearing customer feedback and if we can fix your problems for you, then it makes it better for everyone using the product.
[00:47:34] Victoria Melnikova: Awesome. Thank you so much Paul. It’s a pleasure to have you here and we are super excited about Supabase future. I’m still like ecstatic after the Supabase select, and I hope that lingers not only for me, but for your users.
[00:47:51] Paul Copplestone: Thank you.
[00:47:52] Victoria Melnikova: Thank you for catching yet another episode of Deaf Propulsion Labs. We at Evil Martians transform growth stage [00:48:00] startups into unicorns, build developer tools, and create open source products. If you, a developer tool needs help with product design development or SRE, visit evil martians.com/devtools.
See you in the next.
