Tobi Lütke is still captivated by internet commerce, 20 years later

The best gift in life is finding a beautiful problem that you can never solve. And even if you accidentally solve it, if you're so unfortunate to solve it, hopefully it has like plenty of enlightened problem children. Consumerism is like a thing that like... is being thrown around, but where does it come from? They throw away things because they hate the things that they have. The thing that solves consumerism is quality products.

Personalized ad is a wonderful thing. I am scrolling past something that's monetizing the free application that I appreciate using. You should tell the EU this. Yes, well, I do, actually. Some people fall in love with solutions. Some people fall in love with problems.

I just fundamentally am a person who appreciates people who fall in love with problems. It's very hard to know who is who, because if you don't change anything, they look exactly the same. I'm disappointed you only have half an Irish pub. This is our MVP when we weren't even sure this would be a thing. Toby Luca launched Shopify in 2006. We've been working closely with them for over a decade, and I always enjoy getting together with Toby, and I always learn a lot. Cheers. Cheers.

Okay, many places I could start. What's the size of the code base in as much as, you know, vendor stuff or whatever? I think the sort of core part of Shopify, all the, like, minor stuff like the identity system and all these kind of things is like 20. million lines of code and then

The TypeScript written admin interface is like another I think 8, 9, 10. It's apparently one of the largest TypeScript applications and definitely the largest Ruby applications. We pushed it into pretty uncharted territories for our text. It does so much. It's like a full fintech business in there with capital and these kind of things and all of what we are doing together. And my company adopts the complexity, the real life messy complexities of...

the commerce world into itself to then front it with approachable interfaces, things that don't, you know, just making all of these fit together. And so that requires a lot of work. So there's like... thousands of projects going on at any given time. So going through all of them is work. But we have a good system. We have a product operations team that prepares for it. We have an internal system that's just for the reviews, which is actually kind of cool. What does the system do?

So I'm fascinated with companies. I think companies are underappreciated. We don't think of that generally. Companies are technology themselves. They are technology by which... you create um part of what they create is social acceptance for you know people maybe tens of people hundreds of people thousands tens of thousands people depending um to spend all their day all their day pursuing a mission together you know it's like

It's really just like universities exist giving you time that you can use for like... hopefully intellectual inquiry into some topic and everyone's cool with that although if you try to do the same thing you know just with your laptop on the internet all day long has not nearly the same social acceptance

may also be valuable. And then companies are the ones that allow a lot of people together to try to pursue a mission or create something hopefully world-class. So I think companies are understudied.

Also, the incentive system behind companies tends to be that there's no absolute need for perfecting them there's just like you've got to be better than everyone else and at some point in many industries like if you get ahead it's really really hard for others to catch up and so it's just sort of a point equilibrium at which point you just

Because you're good enough, and so there isn't pressure to get better beyond some points. Yeah, I mean, take an example. I talked with some executives in the gold mining space. It's just really interesting because I'm like, what is your day like? The same question is much more interesting.

It's just like, I mean, my conclusion, which wasn't maybe what he was trying to guide me to, was essentially government relations or investor relations. You know, like, if you're, you know, like, you know how much is in the ground you know sort of it's a capital play but somehow if the investors like you better then you get a higher multiple at which point

it behooves you to purchase of other gold mines really quickly. So I imagine this involves playing golf. That's probably the technology there. Is one of the conundrums for you as someone who thinks about companies and organizations a lot, the lack of

measurability of R&D, which is probably half of what Shopify spends its money on is R&D. And so if you're running a factory, it's pretty easy to measure the inputs and outputs and how many widgets are we producing per hour and what's the labor efficiency and everything like that.

It's very measurable. Whereas when you're building software, it's like, I don't know, we put a bunch of engineers on it and hopefully in like six months they have a great product out there in the market. But the intermediate measure is, yeah, you can measure like... pull requests per engineer per day or whatever, and everyone knows it's like a very broken proxy. I have lots of thoughts on this specifically. I would love to actually know how you think about this too.

So traditionally, when you look back in business history about when people got like a little bit scientific about business or like actually started to improve them, like clearly it all started sort of vibes based. And then, you know, Frederick Taylor came up and, you know. brought the wonderful new technology of a stopwatch to the production line and timed the various steps and of course produced, this is I think, Bethlehem steel and end of 1800s.

drove massive efficiency gains and so that's like the hero's journey of business books right like it's like And there's still so much of the corporate culture today, right? It's how org charts tend to be structured. It's how the financial statements are. The kind of business that financial statements are optimized for is kind of a factory. In a world where almost no company was terribly conscientious.

When no company was metabolized all the value available to them, someone starting to hill climb some source of efficiency gains or just becoming a better company. was like a breathtaking change, right? Like once someone starts hill climbing, everyone has to or gets left behind. So that became the story. And it's remarkable to think that that was probably good enough for like...

probably 80 years or 90 years from this point on, right? It's a lean, basically, which is kind of the 2.0 of that. Exactly. So you said it was sort of a car industry with, exactly, Toyota and so on. new ideas came on, but they were all around the drive for efficiency. The downside of a drive to efficiency is that it requires you to act on quantifiables. You can only measure what you can measure, like, schematically. So that's not the entire space of how to make better.

companies like there's a huge amount of things that are unquantifiable but taste like quality uh is hard to quantify but not only that because those are the obvious examples of unquantifiable but the big one is i'm curious like there's uh

team at Shopify that's doing a great job of their product and there's a team that's lost and has kind of a woolly direction and actually their dev tools suck and so it's very hard for them to make forward progress and they're thrashing if the lots of keep the lights on work just How do you distinguish between the team that's struggling and the team that's executing really well? Yeah, I think that's... I mean, this is why the question in my mouth is...

I should go to, I don't know. Here's what I found books, is have rituals by which we talk every, I mean, at the latest, every eight weeks. And so let the teams talk about the progress against the goal. We have an internal system called GSD Getting Shipped On. It's the central registry clearinghouse.

It's our wiki, it's our feeds and stuff like this, but it also has every project in it. And so it's built around, you know... teams updating everyone else registering like here's something we learned and so like that was step number one let's get all like get the actual state of a business into a legible um internal system

that you can reason about it. Where is every project? What are the deadlines? What has changed? And so this is what these reviews are about. So we go, like, this is what we have on the screen. They get a little TL draw that we can...

drag images and mockups and everything into but that's their area but like all the metrics and all the things are like around the around the edge on the screen right in front of everyone and we can talk about what um that and how many people are on a team and this kind of stuff And giving everyone an opportunity to just talk to me, even if it's only quickly, is incredibly valuable. And I just learned tons of things about it. And there's something that's in the calendar.

I know, it's exhausting and a lot of work, but I don't know how to do it in any other way because that is the thing by which people can say, hey, it feels like progress is really well and here's what we're building and here's something we learned, what should we do about it?

Is it like people who want to get into running the couch to 5K program start with, get off the couch, go walk down the block, go walk down two blocks, and things like that. It turns out that from an organizational point of view,

Having a centralized source where you track all the projects and you list out your goals and you post updates, that sounds too simple to work in the same way that getting off the couch is too simple as a step to running a marathon, but it turns out both are true. I think it's honestly... it's not a complex idea and it's extremely valuable for 15 other reasons other than it powering these reviews.

It's not terribly complex and it works. No, we took inspiration from GSD, your system as we built out, ours similarly, and having a centralized internal source of truth for projects that are going on is... surprisingly helpful. And it feels like it shouldn't be, but it is. Yeah. You know, it's funny because, again, it's an internal system. Shopify has a culture of building internal software.

I tend to point out Shopify itself started out as internal software to power my Snowboard store. So it'd be funny to not do this. And so I think fundamentally, like companies are all...

pretty bad. All the companies of today are pretty bad compared to the companies we will have in the future. And I don't know how I would... possibly run Shopify without having GSD and I think there's going to be some productizations of GSD that'll be available to everyone that's probably going to be renamed 15 times or whatever become a category of software that's going to exist and then

Everyone's going to be wondering, how the hell did we build software before we had it? And I tried to have Shopify live in that particular state a little bit earlier. Has your proclivity for developing internal software ever led in just like a really hilariously overbuilt internal software direction? Oh, yeah, absolutely. Like what's the most ridiculous internally developed piece of software?

You have like a static sort of internal geocities, like it's like static site hosting, which is like it's the most incredible and like geocity. Yeah, like it's like a like 90s web.

um with all the pros and cons yeah like it's it's i mean it sends us down wrong paths like it's it's very often i say internal tools culture this is actually a little bit of a dangerous statement i shouldn't make it like as such we have a very strong appreciation of good software and good in the context of we have to use it.

Some software is really good. Shopify is really, really, really good software if you're running a D2C business or products. Some people try to stretch it to subscription-based groceries. doors, something. Maybe that even works, but I clearly didn't build it with the rest in mind. And I actually, although I appreciate all my customers, I'd rather not anyone use the software if it's bad for them. Because that's like, I build Shopify.

to avoid people having to use shitty software, not to add to the pilot. So sometimes this means, given our scale and our minimum quality bar, that we have to build it ourselves. And in those cases, we do that. build it where we have no business doing it, and that's where things go sometimes wrong. Well, what I think is interesting is you think it's important to build your own HR software, and you would recommend people at least think seriously about that. Maybe you can share the worldview there.

Not sure I exactly recommend it, but we have ideas about how to do HR in terms of compensation and so on that are just different from what's normally implemented. This is most true in project management software, like, again, GST. If you use software by others, you have to buy Interval Vision. Like some vendors...

tell you the software can do absolutely everything. Software have a worldview, so you're adopting Workday's worldview when it comes to your HR, which may or may not be what you want to do. That's right. And I think that's... That's important to realize and it's important to use to the advantage. And I think this is why it's really, really important to buy software. Like the people who buy the software, other people should use the software because they have a better view of what needs to be had.

So in our case, in HRS, we had some things that were not doable with Workday and all the systems, and so we built it ourselves. I mean, this led to a journey that made might have gone a little bit too far on um how should this be done this is one of the areas where like probably Shopify is the weirdest bit of software ever made but like but is it something like The internal systems you use will subtly affect the decisions.

that you make. Everyone's probably been in meetings where it's like, oh, well, we can't do that because the system doesn't support it. And so again, do you want to have a workday-designed compensation system? Probably not. You wouldn't frame it that way.

which is why, again, it's important that people think about, yeah, what is the vision they want and does the software secure? I'm a toolmaker through and through. I have the sort of kits in the LAN parties of the 90s that set up the internet sharing for everyone, right? Like, then that was still difficult.

like I'm a toolmaker, infrastructure thinker in my entire life and I deeply believe in environments that cause people to accomplish bigger and better things than what they even imagined they could have done. Marshall McLaren says, first we make the tools and then they shape us, or something along those lines. I have a terrible memory for exact quotes. But the sentiment is right. And so I think this is...

powerful and is something I want to channel at every level. Again, I want the people who have a utilitarian problem of, hey, I need an online store. to accidentally catch Shopify because it's the go-to software, but then Shopify to inspire them to build much bigger and better companies than they thought they did and like elevate their own ambition in what they are building. Just like my ambition increased from building snowboard stores to helping millions of businesses be built.

It's the environment that I think we have a lot more power over. And it's the sort of missing ingredient people talk about. You know, people talk about incentives shaping companies. and then policies shaping companies. Those tend to be the two tools. I think generally, if people are really insightful, they talk about culture as well.

But it's the environment that's even more powerful than all those things. And it tends to be as a software company, especially if you feel like you have agency over the tools, that you actually have more power over the tools and therefore the environment than you have over policies. because what can be moving forward by the...

by the speed of a deploy. A new change to the GSD software is immediately part of environment where a rollout of a new policy is going to be a conservative town hall and a lot of convincing. actually acts as a fast and the most effective way of evolving a company forward. Having opinionated software that you use to tweak the environment. That's right. So the obvious question is if software

shapes us and kind of shapes the actions of organizations. And you should think about what the vision embedded in software is. What is the vision of Shopify for... your merchants. I mean, you said one part, which is that they should be more ambitious for their business success and they can probably be more successful than they might realize. What else? What I tend to visualize is like, what are the things, you know, if

You're sitting with a friend over drinks in an Irish pub. I'm with you so far. And the question is, like, hey... you've done an online store, you've done an e-commerce business, you've done a retail business for Shopify, would you do this again? But the answer to the question is like, hopefully, hell yes, because... And then there comes a list of things that we would like to do. Amongst those, it just makes it easy. It allows me to build the business not feeling like I have... When you try to...

paint a painting and you have mittens on. There's perfect vision in your mind, but you can't get it out because of the tools. We've all used software like that. Exactly. Another thing is that it just... it keeps me cutting edge. We are going through an enormous platform shift. Like Shopify predates smartphones, like software. So we have seen multiple platform shifts. And usually what happens when you study any kind of... industry upheaval or like a recession or anything it's

hardest on the small businesses. The really big businesses get bailed out anyway. And everyone in between has probably usually the line of credits or whatever capital to make it through. You're talking about an economic recession. Economic recession. So something happens in the world. The smaller businesses, other ones were wiped out. The failure rates during recessions is enormous. And that's a huge drag on economies. Because like...

60% to 80% of all people who work for small businesses. I think people felt this very viscerally in COVID with restaurants, which were obviously particularly hard hit by COVID. But I think people could see how little...

uh how kind of shallow the balance sheets were because entire you know streets turned over with new storefronts like it was a very visual representation for people of how fragile a lot of the small businesses are yeah and like i mean many many places never recovered it's like it's

It's a very organic process by which an area becomes like sort of have the critical mass of interesting stores that like then has people come there and that just like this builds up over very, very long periods of time. Entire cities' centers can shift in these times. And so it's a very precarious balance. And they have very tight margins. There's so much that makes them fragile systematically.

at least all their sales disappearing because suddenly most people purchase from their mobile phones that just their website doesn't work on. It's not their job to stay current with these kind of things. So just inoculating from all these kind of shifts is a really important part of a business. So you're talking about some of the disadvantages that small businesses have. I was realizing recently, it actually really hit me when...

I went to buy something from a very large, famous brand, name redacted to protect the guilty, but very big, very storied. And I was going through the e-commerce experience, and it was horrendous. It was so bad. And I realized that thanks to Shopify, we are now, I don't think it's hyperbole to say we're living in an inverted world where the very large retailers are worse off in their e-commerce experience. They have worse systems and lower converting websites than... small businesses.

which have these, like, amazing, super snappy, and they're just, like, more technically performative. You measure the latency, if you measure the, you know, full funnel conversion, everything like that. On any metric by which, you know, if you were, with your stopwatch and you went from Bethlehem Steel to, you know, the internet, and we're kind of measuring things with your...

stopwatch, the small businesses are better off. So one, I thought that was just interesting that in this case, we have this inversion where on a very tangible technical level, the Rebel Alliance is doing better than the large established companies. But then it made me wonder...

There's kind of a question for Shopify where you have succeeded on the small business side. And I know that's probably anathema to you. It's like, oh, no, we're just getting started, blah, blah, blah. But you have a very significant fraction of the small business e-commerce market with a very successful product there.

And so how do you think about Shopify expansion from here, where you can go international, you can go up market to large companies, you can go into new modalities like agentic commerce. There's many vectors of expansion, but... it does feel like there's a take-stop moment because step one is deliver a great experience for small businesses. Yeah, thank you. By the way, me too. I also have purchased from Nike. It wasn't Nike, but yeah, those kinds of people.

Yes. I love this inversion because this is one of the very few spaces where this is true. Yes, exactly. Exactly what we set out to do. It's actually an access to the capital market. My fundamental belief is people should use amazing software that really, really... perfectly fit to the problem there. You don't blow something. Exactly. But we can go deeper and deeper into the space.

Really, really quick. It's really easy to underestimate the size of... Everyone underestimates the size of the internet and everyone underestimates the size of retail. We sit in the intersection of both those things. My favorite stat is that the companies... The industry supplying wooden pallets to warehouses is a $60 billion a year industry. That gives you a sort of a...

idea for how large retail actually is, at least directionally. And so what we have done from a company perspective is very clear that our mission is to make entrepreneurship simpler and that's where our heart is. many of our customers have gotten quite big. We ended up raising one of our customers to a billion dollars of revenue.

So there's some of our customers, in the top 10 of largest customers, many of them have started on Shopify. The reason why they're still around is because we actually, again, we take this so seriously, that we're actually building with them while they grow. And, you know, it turns out...

again like all these stories of software is for a certain segment of a market are skills issues and that's exactly what they sounded like when i first heard about it it's convenience to the company making the software to say focused on one segment, and by the way, the other segment focused on the richest segment, funnily enough.

And we just said, no, we are starting with SMB, and then some of them grow, and hopefully we build good software, and then we invite the others over. The most common thing that people said with Shopify in the early days was always that this will never work in real life, because in the real world... retailers are messy and you know just like have all these business rules that we didn't implement and our opinion was always like

Real world sounds like a terrible place. We like ours better, and we just invite everyone over. Well, isn't there also something where, like, the real world is messy? But we find this with straight billing, where people say, oh, everyone's billing system is different, and, like, the dimensions of their billing...

that they want to encode and things like that. Therefore, you know, there's a certain size beyond which, you know, strike building will cap out and people need a custom system. And our view is, absolutely, billing is really messy and it's also hard to... test because it involves time travel, right? It's like, you know, what happens if people do a mid-period adjustment that needs to be pro-erasure, whatever? Therefore, do you want to be...

using your homegrown version of this that was developed by some guy who had never built an e-commerce or billing system before and has since left the company. And now it's kind of poorly maintained.

Or do you want to buy it from people who have been thinking about this problem space and building a generally abstracted frame? And, you know, our Stripe building now has a time travel engine where you can just move time around to test all these kind of different cases because obviously you need that.

Shopify's case similarly, yeah, we've been thinking about inventory management for a very long time and, you know, how you model inventory and all these kind of things. And so I feel like it's funny that people sometimes say, oh, this will never work at the large end because our needs are super complex. Therefore, we should be using a piece of software that only has one user and is poorly maintained by one tiny corner of the company. Yeah, that's right. I think...

Look, when I first started programming in my apprenticeship, my master taught me that in the software world you have About two years time, anytime you start an important piece of software to try to nail it, try to nail the problem, because afterwards it just magically, like someone puts them in the code base and you're never going to change a thing. Yes. And that's just like, that's the state of software in the 90s. And frankly, just the way the software looks like...

Everything that got to scale was like finished in the 90s with cement in it and then sold in the 2000s. And in a lot of decision makers' life, that was the time when software adoption really started and their priors are loaded up with software that corresponds to rough.

what Jürgen back then was telling me. Now, we've gotten a lot smarter about building software. We have, you know, CI and automated testing in general, and like, we just, like, we know how to build software in a way. We're by no means perfect at it, and we don't even know how to measure it, as we talked.

about earlier but like state of art has increased significantly and we make better software and that's a completely different thing than what people are comparing it again but it's hard to tell them right like it's you know the first 80 of every e-commerce business are the same It's like you need a checkout, you need credit cards for us, and you're taking credit cards for us. And, you know, like you need the hosting, the checkout, and all this got fixed, and not an interface.

So it's the last 20% that are actually quite unique. And they exist in a particular, very predictable problem domain. Like, what does the website look like? It's like what my brand looks like. Also business rules. We've over time, just like your time-traveling backtest system, figured out where do we allow programming or programming enabled extensibility in the system so that these last 20% can uniquely be created for everyone who wants them. And I think that's what it means.

I think that's a place you can only get to if people work on the problem really care because like you kind of have to fall in love with a problem to want to look at it from so many sides but at some point you understand it well enough to not solve the problems that people state, but build infrastructure in the form of primitives that can be composed to solve all the problems. And that's a completely different level, but it's what I deeply appreciate.

massively admire it every single time any category gets to this quality of decision making and infrastructure. Another thing I enjoy on the software quality side that both Shopify and... stripe sell on, that's just a funny domain to pursue, is peak load, where stripe sells on just...

you can put a lot of transactions through the system at once. And obviously, Shopify, similarly, it feels to me that how you displace other e-commerce systems is if you are getting very spiky sales, you can... put 1,000 requests per second through the system, and that's totally fine. And most systems at a technical level can't do that. I guess I find it funny because I think we kind of grew up together.

kind of expanding the, you know, requests per second thresholds. And you guys pushed us pretty well there. But I'm curious, like... where did this come from in e-commerce culture? Because the marketer is saying, everyone will come to the website and click buy at the same time. The engineers are like, no, please do that. Stop encouraging that. But it's become a real part of e-commerce culture, like the drops and things like that. Maybe you can fill me in on the drop history.

You seem to have started this. It's really funny. I'm just as amused by this as you are. No engineer would ever suggest this. Because it's a very, very strange thing, but it works. fantastic business built around it. The one that did it for us first is like somewhere in 2010. This is called The Jive, which was like some... I mean, it's a website, it's like some community.

some t-shirt with Bill Murray on it. And that thing, every single time I went on sale, I took Shopify down. It was just like unbelievable. We actually couldn't, we didn't know the demand level because we just couldn't. observe it like it's like it's like upstream the routers could not send enough traffic and it's like first of all who's to try if we had to like like catch up very quickly and um so it was one of those things where um

There was a very specific decision moment where we realized, obviously, Shopify down is the worst possible thing. The entire company exists to never make it so that anyone is down, no less the entire thing. We were like, well, the traditional business thing is like, this is super expensive, from a reputation, from money, and we can't even, in our business model, ever make money back from the drive, given the amount of resources we consume.

So we clearly need to fire the customer. And we decided, no, absolutely not. We're using this as a gem. And we're going to build the system in such a way that we can deal with this, because this is exactly what good engineering looks like. That was for Judas, because...

I think the sort of pattern ended up really sticking. It was like Supreme, which really drove us further. The streetwear world. I remember Supreme, and I remember Kylie Jenner. It was like lip kits. I don't really... That's embedded in... my memory from kind of 2013, 2014 Stripe-ish, that was definitely, like, that was a board-level topic. Yes, and we had board-level discussions on our side, too. It actually would be funny to know the diff. You probably sweat it exactly the same.

months on both sides because obviously we have an incredibly long-standing and I think at this point storied partnership. It's tremendously difficult to synthetically test. all the pieces that go into commerce because it's just like so complex with so many things. There's so many APIs. There's so much we have to like the shipping to of course payments and payments is like depending on the mix of bins. It's like

different profile on Lord, right? Kylie Jenner, like the store, had a lot more debit cards, and that just kind of exercised different parts of the infrastructure. And so, anyway, these kind of things you find out. And on our side, no sin was forgiven. Every single sin was revealed and found us to be wanting very publicly.

Is there anything particularly technically hard to scale? Well, I mean, so if you open a book on databases, like sort of page, probably chapter one is like an example of how a transaction works is like... I mean, either moving money on a ledger or moving an item out of inventory into an order. It's a textbook definition of technology. Two textbook definitions in a row.

you are fundamentally have to serialize at those points. This is like the lock contention moment. There's architectures which we employ that you can get around this. There's like pools of... products like we take out what's out of an integer into like a striped thing and like there's all sorts of things that um do and then try to refill this as fast as possible. But the point is that you really deal with log contention and log contention is difficult. It's very, very hard to scale if everyone

at the same time wants to decrement an integer in the database, just actually, then also the transaction involves having to go to interchange and move money. I was reading some histories of Oracle recently, but I think it's kind of interesting because

You know, proprietary databases were the leading databases. Like, the LAMP stack only really took over in the 2000s, but in the 1990s, if you were, like, building a thing, you actually needed to go buy your database, and there was these proprietary database wars in the 70s. And I was wondering, are there lessons for us for the kind of...

current AI wars and model wars from the database history. But it was striking talking about kind of the early feature wars of Oracle versus its competitors and locking. was such a big differentiator in those early versions. And like row-level locking. Row-level locking was a big thing. But it came in like Oracle 6 or something. You know what I mean? You would think it's like an Oracle...

two feature, but no, it took them a few revs to get there, and row-level locking was like the bee's knees. And they released it. I mean, and very glad they did. I mean, yeah, I mean, this is... Dotcom predates Shopify a little bit, but like, yeah, dotcom tech company creation was like, day one, you send a million dollars to Larry to get your Oracle database. And I think, you know, day...

like 180, you go public or something. It's a very different time from today. Yes. I mean, I guess we're kind of getting back to that a little bit in AI where, you know, Step 1 sent $300 billion to Oracle for a data center. You're a couple... days past 180 on your journey to becoming a public company, I have to say. Yes, yes. Anyway, moving on. I want to talk about agentic commerce.

I had a really nice shopping experience recently where I wanted to buy an integrated travel adapter and power brick. And so I did some product research in ChatGPT. And it was really nice. There was no...

ads getting in the way or anything like that. I actually found a product that I hadn't come across from Momax. It was really good. I think I was telling you that. Yeah, I had to. Okay, yeah. And what's the verdict? It's very good. It's very good. Okay, yeah, we'll put a link. And it kind of linked me out and I purchased, obviously, a super easy one.

like with Shop A and everything like that. And it felt very futuristic to me, both in the product discovery and the completion side of things. But it feels like filling out... Web forms is not a value add activity.

for people. Like when you travel, you probably don't directly book the hotel yourself on the website. You know, you have an assistant who does it or something like that because choosing the hotel might be desirable for you, but actually filling out the web form is not a value-add activity for you. And so...

What is your vision for how agentic commerce happens specifically? Because it seems relevant to your interests. It's relevant to my interests. I mean, it will be extremely valuable, and it will be something that... will be done a lot and might even be a majority of commerce on the internet, I don't know. But I think my role in this is going to be largely infrastructure providing.

Our job is to keep all the millions of businesses of Shopify extremely current. And so what we want to do is make sure that everyone's plugged into the various chat systems and this. an MCP connected to every Shopify store and a global catalog that's really nicely presented. We're building software specifically for

the open AIs and the clouds and the complexities to make it really, really easy for these products to show up and reason about them and so on, because we think it's really, really important that they can show up. So there's an infrastructure angle in it. how it's being used. So I think the best way to predict the future in most cases is just like, frankly, look at what rich people buy for themselves, right? Like, you know, Uber is wonderful.

company is also extremely predictable because rich people had car service and you know so it's figuring out a way to scale that in such a way from cost basis to so that everyone can use it it's just like yes that works so uh personal shoppers right these are the places where uh you know memory and understanding people and personalization are just like pay so many dividends and are so clearly auto good like it's like

People have to make so much out of data and advertising. You know, a personalized ad is a wonderful thing. If a personal ad tells, I am scrolling past something that's monetizing the free application that I appreciate using. You should tell the EU this. Yes, well, I do, actually. And the fact that it's highly personalized to me means, like, at this moment where this platform monetizes because they're giving me a gift that I get to use.

it's win-win it's it's win-win using this moment also that is now not just like a ford 150 truck or i don't even know if that's how you say that but like it's like i'm not a market for trucks right so Instead, I see a travel adapter, like what you just sent me. I am your personal shopper, in this case. But the funny thing is, you sent me the link, and if I would... put that link in ChatGPT, which has lots of my history, and ask,

Is that the kind of product that I would appreciate? It would just return yes. Because it solves the problem in its entirety. It's like a travel adapter, which for any country, to any club, to any other, all sorts of stuff, good product. I want that. Okay, so we're getting down the road here. I mean, it feels like step one that's going to happen with agentic commerce is I feel like existing aggregators, if just aggregation, is their raison d'etre.

then they're going to have a lot of questions that they need to answer because ChatGPT and the AI apps end up as a new aggregation point. it seems to very much benefit the tail. It benefits all the Shopify merchants, just like, you know, we had a new set of content creators emerge on YouTube that effectively compete with kind of TV. They're both just...

entertainment, and the tail becomes much more powerful when you have over-the-top distribution through YouTube. Similarly, it feels like once you have over-the-top distribution through AI apps, it becomes

You have a much more powerful position as a tail merchant because, you know, Momax is a tiny brand, but they can just get recommended as the best product. So that's step one that I think happens. I think a slight... modification i would just make this i don't think the tail part matters i think merit matters it's like it's just like it's it's kind of an important point because like you know often when you talk a lot about commerce and products at some point just because of

prior discussions, people are like, yeah, but aren't you just feeding consumerism? So this is sort of like the negative take, right? And again, consumerism is like a thing that is being thrown around but where does it come from it comes from people throwing away a lot of things that they probably shouldn't but they throw away things because that they hate the things they

They have. These things don't work. They fall apart. They don't do the thing. They're like cheap, plasticky, this kind of thing. The thing that... Solves consumerism is quality products which you want to keep using, right? Like you want to be excited about this and like you want this travel to last forever. And so you never have to think of, you literally never have to Google what plug.

It's been used in Singapore. You just know that the thing that you have with you can deal with everything to everything. So problem solved. I think that's an important modification on the point. Yeah, I guess implicit in my view is that there's good stuff in the tale.

Whereas the head has distributional advantages, and so it is more meritocratic to bubble more stuff up from the tail. That is, I think, the reconciliation of those two worldviews. Okay, so that's kind of step one that happens. I think you are getting to step two, which is... It feels like there's a lot of changes.

potentially that can happen around personalization, about agents actually acting for you. And maybe one thing, I'm curious if you've thoughts on it. It just seems like a lot of search in e-commerce is really bad, where it's keyword-based search still. And you can't do things like, let me know if this becomes available in gray. Or what other products do you have like this? And you want some kind of embedding space.

that thinks about similarity between the products, and no one can do this. So when are we getting better... personalize it and search. I just want a really good search of the products that are out there. I am obsessing about this now, and we should have been- Tell the secret Shopify roadmap. Shopify should have solved this earlier and hasn't, and I'm pretty- mopey about this. You guys are okay. Yeah. I think we will solve this now.

It's really, really fascinating. I really deeply appreciate it as a field. It's so different. It has its own lore. Everything is interesting, is my fundamental opinion. And search has... been a wonderful area to like actually learn a lot more about the history now and like the fundamental thing is like there's a generic bias in search and a text is king and like Frankly, the people who worked in search are fundamentally more interested in searching through.

papers than for products yeah and so there's very few of the best people in search in in a world of people who know most have actually ended up working specifically on product search which is really a should be seen as its own completely different domain that just sort of looks like it, but it's really clearly deeper with lots and lots of complex features and so on. So I got to build up over last year.

really, really, really, really great search team and exactly look at the embeddings and these kind of things. The amount of improvements that it's just like. Unexplored is just staggering there. Shouldn't you do search across that? I mean, I find it interesting when Shopify does things broader than the Shopify universe, because the Shop app can track packages not...

bought from Shopify merchants, right? Yeah. And so shouldn't you do the search thing across the whole web and not just Shopify? No, should we? I think so, because the thing that feels like the big opportunity to me is

clustering was always so hard to do because all the features were developed manually. And so you had to go say that the dress is red and everything like that. And it feels like, have you ever used LLM as just a recommender engine where I like feeding in, these are all the books that I've read recently. Give me a narrative nonfiction book that someone who's read all these books might like. And it does extremely well, but you're kind of using the LLM to access its world model.

It feels like now you can have an interesting product world model and access it, but people... like for the people using it, whether a merchant is on Shopify or not is probably an incidental detail they don't care about. Yeah, I totally agree. And you're talking about how, but like what is the really important thing? I just want the LLM to be proactive and just tell me, hey, there's like...

this outfit would look good on you and here's a preview. And just like, here's a full costume you wanted. It's the personal shopper outreach kind of component, which I think is just really valuable. And this is not going to be some kind of... You'll get this from everyone because it's really easy to put together. It's hard to get the data and the embeddings and the sort of understanding about, you know, like we have to figure... It's funny.

The most important thing, again, in the AI space is also the unquantifiable. It's the vibes, people call it. It seems like Midjourney specifically is one company that managed to actually scale taste. A very distinct aesthetic in the image. Yeah, and maybe this is because they drove towards a particular aesthetic that they just sort of kind of felt they had ownership over, but it's one that's clearly appreciated. But David is also like deep founder mode type how he operates. Yeah, exactly.

When something outperforms, it's usually the product of a place that doesn't make decisions by consensus, as we know. We need to figure out taste in this product sense. Yes. We'll figure this out. Someone's going to figure it out. Maybe everyone's going to figure it out. Maybe it's already figured out in the neurons, and we just don't give enough information for it to reason around the taste.

I think it will be very, very, very good personal chopper. The thing that I think the aggregators still, to just get back to your point on the aggregation side, there's many reasons why people buy things. Lots of things are input purchases. It's like the things next to the checkout counter at the supermarket. I don't think people are going to use

chat GPT to buy mass bars or something, or candy or so. You don't do so online. Right, exactly. But is there an online impulse purchase that is not suited to happening in an AI app? Interesting question. nothing comes to mind immediately. I mean, I guess online impulse purchases are like things where you're scrolling through Instagram and it's like, you know, all the COVID stuff. It's the sharpest knife ever developed. It's an umbrella that changes color in the rain.

I wonder if those are truly input purchases. I mean, they sort of feel like input purchases, but they tend to be, like, very few people part with money unless they are, like... holy shit, this is the thing I always wanted. Like this is actually usually the longest. It's just tapping into demand that's been there for a long time. It's sometimes a little bit, the opinion, when you look at log files, you're going to get a very incorrect view of the world.

it's very often the longest consideration phases. For years, I wanted a travel adapter that can deal with everything to everything. It goes in a chat from you or in an Instagram ad. It doesn't matter. I'll insta-purchase it at this point. Okay, you were just ready to do that. My other purchase I made recently that I really liked is...

I was going on a long day hike. And if you read the accident reports, it's often people day hiking who get into trouble because people who are going out on a multi-day hike, they have their tent with them. They're prepared, you know what I mean? And then people go... It's a whole problem with bicycle helmets too, right?

Same issue. It's a short journey, exactly, just to the store that gets people in trouble. Yeah, so day hikers, they get in trouble because they think they don't need any preparedness, and then they end up out after dark and, you know, a rescue party and everything like that. And so it was like, it's called bivvy, you know, like a little sleeping bag for... emergency purposes. And it's made out of the reflective foil. And so again, it's if you...

end up stuck in an emergency foil sleeping bag just so you don't die of exposure. But it folds up to it's like that size and so you can just stick it in a day pack and now you are somewhat more prepared for the elements than you were before. I don't know, do you have a favorite recent purchase? I like the... I just really like this... I mean, someone going just like to ridiculously further and to just like almost...

like products that just celebrate craftsmanship in a way, even in places where just like no one else does. It's like, weird example, but like I, you know, take all these vitamins in the morning and like I've bought these like... CVS vitamin pill cases at some point, just for travel. And then I found on Shopify this Japanese perfectly CNC milled, unbelievably beautiful pill cases. The entire website is just like a love letter to precision engineered.

like tolerances of sub-millimeters. What's it made of? Like metal, which is like absolutely, it feels like, it feels so, whatever. I'm really excited about it. Let's put something. Put a link in. I forgot the name, sadly. And now I gift those things because they're just beautiful. Yeah, yeah, yeah. That's cool. How's Shopee doing? We referenced it there.

ShopBay. ShopBay is awesome. It started out this life as a feature called Remember Me, which is literally what was on the checkbox. We also had a thing that was called Remember Me. That's like the original name for Link. No way. Yeah. That's trippy. That's very funny. So, which, you know, it's like, I mean. It makes sense. They're kind of the same thing. I mean, shop is much broader, but they're similar things. Yeah.

Actually, Shopify feels more narrow. It's like it's on commerce specifically. You guys do all the shipping and tracking. Yeah, that's right. I mean, Shopify is Shopify's brand for buyers. Shopify is what you use to be on shop. And this is what we think about it. So, you know, there's a shop app and, you know, shop pay. So, I mean, again, it started because it just is a ridiculous thing that we asked, anyone asked.

anyone to fill out their address with their thumbs when you purchase something, which is insane. This was meant to never be a thing because, you know, like in the... I don't know when they came up with HTTP, but there's the... 402 error code, payment required, right? It's like built into the HTTP standard and no one ever implemented it. And so moving money around was supposed to be built into the platform and it wasn't.

Credit card forms were hilariously insecure over a normal weapon when we came up with HTTPS. reasonable, I suppose, and then it became good enough. There's all these efforts, Apple Pay, I think actually PayPal was around, but then sort of didn't do anything. You know, so there's so many people doing this. okay well clearly this is going to be solved soon but like might as well like just put a remember me thing in there so we can solve this now

as almost a polyfill for the future of the internet. And then no one did. I mean, very few people did, and it just sort of didn't happen. And then now ShopPay is like incredible. Super glad to have it. It's like the most converting thing we've ever done. Yeah, what seems interesting to me about it is... I think it's a significant part. It's not the only part, but it's a significant part of the inversion where, again, now the...

small stores are significantly better off than the large stores online. And that's just a striking... It's striking. And it's a 16% conversion increase by it being robbed.

So it's a remarkable product now partly because, again, it's a power of branding. It's purple always and people learn to trust it and associate it with a store that's actually recent and cared for and doesn't... you know if they know they don't have to before they get to checkout they know they will not have to type in the address with their thumbs like some kind of

Heathen from the 90s. Yes. You mentioned personalized advertising and being a big fan of it. It seems to me that there is a real symbiotic relationship between Instagram where people are kind of scrolling in an aesthetic mood looking at nice things and then... Shopify stores where they need customers. Especially during the pandemic, there was this huge growth of direct-to-consumer brands who really cracked the...

Instagram, CAC to LTV math, and they were able to grow in a big way there. Then, of course, the Apple ATT changes happened where for a brief period, it got harder. It seems like Meta has mostly solved that. But anyway, I'm just curious, what's the current state of, like, is this worldview I have true that there's a lot of Shopify merchants that have been able to really scale up thanks to...

personalized direct advertising? Yeah, 100%. It's a main growth channel is advertising on the platform. The Meta platform specifically is just so incredible for this. the the the meta shopify lau has created more businesses than like uh um you know any government policy in history like it's like so it's like I mean, if you just compare to television and consumer packaged goods before, the television channel

required broadcast. And so the only things you could really advertise were like mass market high margin products. Yeah, yeah, yeah. Washing powder and all this kind of stuff. Exactly. So it's like, you know, laundry detergent and toothpaste. Yeah. And so... The Instagram, Shopify is a precise inversion of this. It's like highly, highly meritful products. It's the first time you can advertise niche products. Exactly. And the niche product is actually doing better. That's the crazy thing.

Advertising directly snowboards is like impossible there. Kind of like a billboard for snowboards. The brand halo of Burton means you can't compete with that. There's intersections for like snowboards. it's only for powder snow on the U.S., West Coast, and in certain environments. This is not the best example. Usually, there's three different intersections of something. And it goes back to Kevin Campbell.

Ali wrote about this in the early 2000s. He wrote an essay of a thousand true fans and the internet is about finding your thousand true fans and everything else will happen. restoring that idea or bringing this idea into retail has done a lot of good things, right? Like it just created better products, it created visibiosis.

like meta allowing you to discover one of those people who has a potential to be one of those 1002 fans is an incredibly powerful thing that just like again led to enormous good and employment and all these kind of things. So it's a very, very powerful combination.

There's all sorts of other ways to help people grow, but this is a very, very large part of everyone's strategy. You're saying this enables 4K for commerce. There's this higher resolution, more specialization, as opposed to the blurry generic...

Yeah, exactly. Find the people who are really interested in, who appreciate the products. I like that. How is the stablecoin rollout going on Shopify? We're obviously doing lots of stablecoin stuff with lots of people, including you guys, but I feel like...

You guys are one of the biggest bets on stablecoin pay-ins. Everyone should just be able to use stablecoins as a payment method that we've seen. And so, how's that going? I'm a huge fan of the concept of stablecoins. I love the work you're doing there and congrats on the launch of Turbo.

And the internet just needs its own currencies. I think it's important. At least it needs its own infrastructure to move things by internet speeds. And I'm excited for it because, I mean, I think a valid criticism of crypto in the past, as exciting as it is, has been that the utilitarian difference between a real US dollar and a...

US dollar crypto coin it's like That the US dollar crypto coin is mostly like it's sort of like gift certificates to scams It's like there's not that many things you can actually purchase with it that are of high utility value. It's like, obviously, this is, I think... The economy was too circular. Exactly. Most of it was like you purchase other coins that then you might have...

I mean, hopefully you diamond handed into some kind of upswing, but that's not usually what happened. And I think that's valid criticism. I think the most important thing to happen there is we need to expand. what can be purchased with stablecoins to the things people want to purchase, like outside of the sort of insular crypto fintech.

working hard on getting all Shopify stores to just take... What does the merchant response mean? Very positive. Again, because through the work that British is doing, we can settle into US dollars directly. So the merchants... They've just gained a new ability. without having to make a strategic choice to go into a new industry. There's demand for being able to spend stablecoins and they can now sell to everyone who wants to particularly do that. So it's totally transparent.

beautiful thing, amortizing R&D over millions of businesses with us centrally because we can just build up the system. And everyone gets benefit of it. You know, obviously, it's in testing. The conversion rates have to not be negatively affected, obviously, by its presence. So there's some, like, we have to make absolutely sure this is the case and, like, lots of buttons of caution on every world record from the checkout because it's like the...

Probably, I mean, it's the busiest checkout on the internet, basically. So we might not want to. reduce conversion rates ever. I think it'll also be interesting to watch the consumer training aspect of it, where QR codes is always my example of this, where the tech has existed since the 90s, but it was really only during the pandemic when we installed all the software in people's brains to understand how QR codes work.

But there was a moment, and now they're kind of everywhere for everything. And obviously, they're in Asia before they're in the US and everything like that. I think there'll be something similar with stablecoins where they'll be offered at the point of checkout way before people become super familiar.

But we're probably only a small number of years away from some mass markets. I think it's going to be very, very natural for people. But also in a way that they just not think about it. It's like it's a normal thing. It doesn't matter. It's going to be so well tooled up to bridge between these things that it's using crypto rails while using the internet and until you need it out of the internet, it's not like...

You don't have to bridge anything. Yes, yes. You referenced a few times the kind of stuff Stripe and Shopify are doing together. And, you know, we've caught up at various points and even talked at the meta level about the Stripe-Shopify relationship where there's some...

I don't know, is it like Apple TSMC or something there where it's like deep intertwined building in a stable way, but also kind of expanding over time. And no confusion as to who's ultimately the one building the end product versus who's kind of providing some of the underlying infrastructure. But I'm curious what you think has worked well, because, yeah, a lot of...

corporate relationships do not last. I mean, I guess if you dated 2012, which I think is accurate, maybe 2011. 2012, like 13 years, it's certainly more tenured than some. Yeah, and I think... It's almost more impressive when you think about it as market cap expansion. People get greedy. We have had a highly functioning, excellent symbiotic partnership.

And probably there's like, I mean, how many more digits on the market cap of both companies? Like we've expanded many, many, many times here, many out of magnitude. And that's pretty rare because like, again, partnerships, especially in tech. Tech is extraordinarily bad at partnerships. It's always treated as a

prisoner's dilemma that at some point, someone in the company realizes defecting gives you five points on the math, and that sounds really, really good with the short-term focus. So everyone must trust each other, and it's fairly rare, and I think it's cool that we pulled it off. And I appreciate the partnership. It's been really, really fantastic being able to just focus on my space instead of having to duplicate.

things that Stripe would always have done better. I think a lot about main quests and side quests. I have such good, I guess, founder market fit with commerce. I really care about this and I really, I have, I can spend my entire, like it's one, Karl Popper calls it finding, like, the best gift in life is finding a beautiful problem that you can never solve. And even if you accidentally solve it, if you're so unfortunate to solve it, hopefully it has plenty of...

enlightened problem children that you can then tackle. So finding one of us is good and not having to meander into some complete adjacency to support it is actually wonderful. Yeah, I think there's also something about, when you talk about this idea of focus, I think second acts are kind of a bit over-valorized in Silicon Valley somehow.

where everyone talks about, oh, we want to do like a new AWS is kind of the coolest thing ever where you do a second thing that is completely unrelated to the first thing versus NVIDIA is the largest company in the world. And they just started making GPUs in the 90s. And it turns out a lot of people want GPUs. And I mean, they did obviously lots of incremental expansion from there, like calling them GPUs is doing a disservice.

but it's in a very hill-climbing expansion way. They went vertical. I think it said horizontal. Business books tell you to go horizontal because, again, the... chief competency of most businesses ended up having a stopwatch. It was sort of a management system world, but we are actually in a different world now. It's like this is commoditized. It's easy to access and probably GPT-5 is better at it than most managers.

And it's now actually like, well, what problem do you understand better than what's in the training tokens for GPT-5? And what do care more, which is a human capacity specifically? Jensen just cared more about GPUs and saw more potential in them. Isn't there a bit of MBA-ism here too, where if one was trying to, when Shopify was a thousand times smaller, raise money for Shopify, it would just sound ludicrous to say,

We're just going to do e-commerce software. And it turns out that's a really big market. And here's our current projection. And here's when it's a thousand times larger. It's like, get out of here. Come on. Whereas if you say, oh, we'll expand to do this and we'll expand to do that. We'll build a CRM or whatever. It's easier to justify, you know.

expanding into other terms. Like I got rejected from plenty of VCs in 2008 when I went around and some of them said they would invest and then it was usually terrible terms of like ratchets or multiple. And I had to move the company to Silicon Valley, which again, I would if I wouldn't have told me to do it. Like once you tell me to do something I don't want it anymore.

but anyway so like like and i caught up with many of the partners later in one case a really interesting conversation it's just like he asked me what what did we miss back then and i just like well the reason why you didn't invest is like like you said like with 40,000 online stores in the world right now in 2008. And well, you can get half of them, but that's not actually that big of a business. And what we miss is like, well.

Shopify itself was the solution to that problem. The reason why no one was building online stores is because the friction was so high. And so it's a little bit of that, right?

Yes, it's not an interesting business plan. The pitch isn't very good. Specifically coming from me. This, but more. Yeah, coming from me is probably especially bad. But it's also kind of, man, the best products in the world come from people just... like really really really really uh being deep on something like in a programming world we are using cursor which is basically vs code vs code is eric gamma's fourth editor

It's like he spent his entire life making these things. It's like the craftsmanship. Yeah, it's like he wrote the design patterns book that many of us read at some point, literally about programming text editors. It's like, yeah, sometimes you just need to dedicate a career to a field and get deeper into it than anyone else and celebrate craftsmanship along the way and kind of try to build the best thing. Yeah. But again, it's hard for people to give credit to that idea and the modeler.

People are not capable of being sufficiently optimistic about a business that really works. It just blows through all their projections. Yeah. The question is also, like, we've seen the story so many times that, like, some step function changes. Like, at some point, like, maybe the shop pay and, like...

like all these kind of things with Shopify, it somehow became something very different than what it otherwise sort of, what you would have visualized. And I mean, this is in no area more true than in the NVIDIA case where like, I mean, these CUDA cores, the Tensor cores, were around. They were on these cards which everyone put. They were part of the bomb.

and very little use by any of engine. Like, CarMax's new engine probably used them for something useful, but they were just there. And Cuda was not even, like, that was around for 10 years before AlexNet got trained. It's an incredible vision to doggedly pursue this because in the future there will be value if you do this. It makes you wonder. I think that's probably true for almost every space.

driven to that level could become so big. Isn't there something here where more founders should be willing to follow their nose of what's just a good product? Because, again, I think that what can happen as companies scale is... Founders get successful by building good product. And then they hire a bunch of people and finance people and run the planning process and everything like that. And it's like, OK, well, what's the revenue projection for this? And things like that.

Stripe Atlas for incorporation. There was no... amazing projections behind that, that one day this will be a giant business. It's not a giant business for us by any kind of revenue metric, but we think it's very valuable because people start their business on Stripe and we help them with something foundational, like incorporating the company and get it off the ground. And it all kind of ties together.

in this very immeasurable, nebulous way. And kind of like you're saying with CUDA, I presume that it's hard to justify at that moment in time, but it seems like the right technical decision. And there's something maybe where founders should be a bit more willing to... follow their noses on what is right for the product and right for the user and it'll all come out in the wash in years down the line. Yeah, I'm a great believer that if you build things that are of value, then...

people will figure that out. It's very, very hard to keep people away from things that make their lives significantly easier, better, or more joyful. I mean, I get the front row seat to this on my platform. It's crazy how fast some success stories. One of my favorite things we do is we started making this really sort of nice...

awards for people when they hit certain sales milestones. And, you know, some of those, when I go travel, I check out who's just... got them and just like sometimes deliver them and Harley does this much more than me and like it's it's such a cool thing to like to talk with them and man like some of these businesses that had like a hundred thousand or million sales are like

18 months old it's just like just cracked the code or something of and and and with the most you know talk of a thousand true fans it's like it's like millions it's like an army created because people just were starved for This like authentic, like unabashed, like something needed to be done. Something was wrong in the world. It involved a product not existing.

And that product not existing was such a bother to someone that they made it their mission to create it. And then they went through trials and tribulations, usually... they are often documenting the journey, allowing other people to ride shotgun to, over time, try on their vision of the world, which, again, is not to change the world, but it's to add the thing that was missing.

It makes things 1% better, not like sort of everything is new. And they just agree with that and they love authenticity of it. They love that someone that they know have a little bit of a... relationship with like kind of pursuit this thing that they think it was worth doing and then the product comes out and and they tell everyone and then more people buy it

And it's just like there's so much good that comes out of this. And that was all value, which was there for everyone to take, right? Like every product that sells well was a discovery. It's like it was sort of invented, but also kind of discovered. But it's that person that did it, and they deserve the economic value that comes to them. Everything about this, it's the best part of capitalism, all rolled up for everyone to inspect and review. And we get to have a front row seat from data and...

proceeds sometimes with giving these sort of celebration artifacts. And anyway, so this is always the greatest thing. It's just like the people who are using the platforms are actually really inspiring people. Well, you know... I feel like it must be particularly satisfying for you because there's all these edge case failure modes of capitalism of, you know, rent extraction or oligopoly or regulatory capture, everything like that. And you just have a front row seat to the pure competitive...

differentiation, people succeeding on the merits, and it's kind of a very pure form. I mean, I feel like you have that too. And it's like... We should write postcards to the rest of it. There's something really to this. This real world thing. The one everyone says, your product doesn't... wouldn't work in like that sounds so dystopian is like it does really sound quite dystopian like so i i don't want to have anything to do with it i just much rather like

hey, just build this whatever world, whatever that is, and invite everyone over. And if it's good and works, then people will join. It's actually that simple. And I think it's better to live in a world with not... like people going direct uh people communicating right people being real people it's yeah there's no there's very very rarely it's very uh pr department or like even there's probably a marketing strategy but the marketing strategy works like based on

you know, just like new ideas and organically. And with very few people being highly enabled, building products that they care about. We have McKenzie from Ambrook on here, which is SaaS for financial infrastructure for...

But she was saying one of the big trends in agriculture right now is direct-to-consumer stuff, where you're actually transacting directly with your farmer. And I'm sure a lot of them are running on Shopify as well. What have you learned on the Coinbase board? I never had another job, really.

my apprenticeship and mostly learned what not to do as a company and then sort of had the opportunity to start Shopify just trying. literally the opposite, and that's basically the operating system I've been running on. It's been amazingly successful, so thank you Siemens. But it's so hard, especially building it in Canada, like in Ottawa and Toronto. It's funny, I always had this effect of I came to value a lot when I had time and money.

I had all these meetings and then talked with people and they described to me how real companies do these things and then I went home and tried to implement it. And only years later did I realize very few other companies describing what they're doing actually.

did the things that they told me. Everyone gave me sort of idealized highlight wheel and I compared that to my average and then I implemented how i think you could do a better job on the stuff that described and sometimes ended up with good things and i was very happy with this and then now obviously i'm describing when this goes really right And just as often, this went really wrong. Shopify ended up being this lopsided thing that is probably ludicrously...

over-engineered but advanced at certain things and then incredibly primitive at things that no one fails usually. Because in a more liquid labor market where more people build companies. the information is just more readily available. What was awesome was joining Coinbase as a board because it's like, here's another founder-run company by an incredible entrepreneur that's put together.

not by committee, but by a clear desire to build an important company, an important institution. And just holding it up and seeing what they were incredibly strong at.

compared to like like it gave me it i learned so much about my own company and i could like you know reference point have a reference point right like so it's it's it's quite a gift i i i don't know when it's the right time for an um successful entrepreneur to join another company I wouldn't do it too early but like it eventually becomes quite valuable and so Coinbase is such a remarkable company you had Brian on he's such an incredible entrepreneur

The very first principles. And there was a company under brutal attack by the regulators in a way that very few companies of all time have ever faced. The government wanted them dead. pretty much settle right so i learned so much about the posture that uh brian took to it he will never get enough credit for the yes for the courage and like the the intestinal fortitude is shown there

because many of the details are internal to the company and the board, but like, God damn it, he's like good. I do not think I could have navigated my company through a similar barrage. Yeah, Brian strikes me as someone who is very hard to manage up to. You know, that's always the risk for CEOs is that people are trying to, you know, construct some Potemkin village. And, you know, he was describing the policy.

uh work that they were doing and he was saying you know well um you know the company's meant to be the good cop and the trade group will be the bad cop and then he's being the trade group and it's like wait these are also the good cops and so he was saying okay here's our new bad cop plan but again he was just kind of refusing to go along with the

the things that everyone was serving up to him as the plan. I think great entrepreneurs are ungovernable. And the problems often arise when people somehow manage to figure out how to... I think studying failed companies is actually often more useful. What you find, invariably, is that the company knew exactly the stuff that was the problem. It's just like, somehow...

CEOs got disconnected from that. Yeah, it's not so much that I want to study failures. It's that I really want to avoid making the same mistakes that other people. have well-intentionally made and, like, found, oh, yeah, no, that's a blind canyon up there. We should at least be able to make new exciting mistakes. Rob, the guy here who runs CorpDev, you know, he's very experienced in it and he's done hundreds of acquisitions, you know, previously.

but it makes him way better as a corp dev lever because it turns out that there's lots of ways for deals to go sideways. And if you just have seen a lot of them before, you can... keep tabs on them and be finding ways to steer away from them, things like that. And I think just...

avoiding the well-trodden mistakes, and if you're going to make mistakes, at least make new mistakes that haven't been discovered yet, is kind of underrated as a strategy. It's totally underrated, but completely vital, because you died to a thousand paper cuts as a company. of have to like first avoid like it becoming a thousand and then second you will make mistakes make them original so there's alpha in it so that you accrue some additional knowledge which now is unique to you

How do you like to be ungovernable? How do you like to avoid being managed up to? You probably can't share all your secrets, because part of them are part of the special time. I just think going direct a lot inside of a company, too. The tourney of the org chart and information flowing up.

polished at each step yeah i'm just like i i don't i don't think like up and down is sort of not the right mental model it's in now it is good it's like uh like certain decisions are best made at the outskirts of a company closest to the customers and um

But I'm a big believer in many decisions have to be brought inwards as well. This is why product reviews are very centralized. It's like a way of efficiently moving information completely independent of org charts. I'm a fan of functionally organized... it's the trade-offs it's like it's

It is significantly harder for executives. It's very unpopular for executives because obviously they just don't own the entire shebang that they can make all decisions in, and that requires a lot more collaboration. It feels wrong and feels slow, but you make vastly better products in functional organizations, at least for Shopify, and that helps.

I just know a lot about the company, being in so many details. I try to be in AI stuff first so that I can... I always see it as a gift when something new happens because... I'm very good at quickly figuring out stuff and then I sit there with knowledge and I can now look at who else gets there fast and those are my future leaders.

Like that's a very implicit test. Oh, that's interesting. You're kind of testing some adaptability or something like that. Yeah. Like the biggest difference between people inside of any company or just like in general in industry is like...

Some people fall in love with solutions, some people fall in love with problems. And I just fundamentally am a person who appreciates people who fall in love with problems. It's very hard to distinguish during fairly stable times. It's very hard to know who is who. Well, if you don't change anything, they look exactly the same. I like inducing change in general to suss this out. I like... I don't know.

How do you induce change? That sounds euphemistic. I don't know, just like literally like that, like do stuff. I think reorgs are sort of a standard way. All reorgs work because reorging is valuable. It doesn't matter what your organization structure is afterwards. There's some benefits to any organization structure. It shakes things out of the stasis. Well, and the advantages of any organization structure occur immediately. The disadvantages only appear over time. So if you change a lot.

you do multiple things. I think heightening your misalignment with people you don't want in a company is an enormously important thing. So if you reorg a lot, you end up... The people who have stepped out who don't like that. It's probably the same position to take even. If a company thrives under a lot of change and is exposing itself to a lot of change, then it's important that you are clear about that.

I think you try to make your company as different from every other company as possible, which is actually, I think, the opposite of what most companies, most CEOs optimize for. And probably most professional management. Right. I think a lot of people want to make their company

as similar as possible to all other companies. But to me, that's a lack of diversity. There should be diversity in companies because there is diversity in people and people need to find a product. In what ways is Shopify most different to other companies? I mean, the internal tool culture is maybe an example of this. It's quite clearly R&D run in a way that quite a few companies are, but they don't acknowledge, they're obscured. We actually say it.

celebrate it and say we actually are on our opinion at this point that everyone should be able to code at least sort of since vibe coding is around this is clearly this has gone from a you know have to spend all of your teenage years in a dark room cultivating a rare skill to spending a couple of weekends with YouTube. Like, let's go. This is not a problem anymore. Speaking of teenage years, I'm always very interested in the on-ramp to

entrepreneurship and the on-ramps to programming. And I'm curious, some people decry that programming has gotten harder to on-ramp into, where, you know, the first stuff I built was like janky PHP webpages, but obviously PHP is actually... very nice because, you know, you're just writing code directly in the web page and you refresh it. That's the execution environment. And now... It's a high watermark of our industry. And so have we made programming too hard to...

get into? Totally. Okay. To tell my wonderful board member, David Heinemar Hansen, the high watermark of the world of programming, at least web programming, was clearly when the... ftp php file up and it was instantaneously deployed how long does it take code to deploy these days like this is like all choices now right like i mean we've made some choices which are worth making maybe we'll do it again yes but like

There was an immediacy and a visceralness to programming, and that clearly has been lost. So lots of things have gotten incredibly amazing, and I always said, like we used to say that... After two years, it's like someone puts them into court base and no one changes anything. We figured out how to avoid those horrors. I think we have a very mature discipline. We've always had like a identity crisis. We always wanted to be someone else.

We have significant science envy. We even attempt to call it computer science. There are some people who do computer science in the machine learning world, so it's a real thing, but clearly that's not. It's a vocational art. It's not most computer science degrees. Yeah. And this is also how we eliminated things like aesthetics and beauty and stuff from our code bases. So I like it from talking about it. There are beautiful code bases. There's beautiful code. It's like this code that like...

tells a story and clearly communicates what it's there for to everyone who reads it. Ideally, it's an artwork of sorts. And so, yes, I think we have progressed. I think we have also advanced, but I think we don't have a clear vision to drive simplicity. And I think I personally... really deeply appreciate and have always prioritized hiring electrical engineers into programming roles just because they just have a better idea for the costs of a deficiency. Waste, really.

like you put too many chips on a thing, it's too expensive to manufacture, you will just never sell it, it's not a good product. The zero marginal cost of copying software just means all synths are forgiven. And in fact, of course, still runs and does everything, right? You really have to fit the whole system in your head as an electrical engineer. That's right. And so anyway, so I think the immature...

We don't know what great is. We leave a lot of this up to people and therefore the outcomes are so different depending on who is on the project. I mean, again, we are all getting more conscientious about it and I think we are in a counter swing. to the horrors of architecture, astronauts that have caused a lot of damage to our industry over the last, especially in the 2010s. So we are getting in a better spot.

I don't know what AI it was. But the interesting thing is that I don't even think it's going to end up mattering because AI is going to solve all technical debt. So it's like maybe all it's in- Same question about on-ramps to entrepreneurship. As a kid, you were selling things to your classmates. And what I observe is a lot of entrepreneurs have the same origin stories where they were...

entrepreneurial early, either something like that where they had some business in school. A super common one is Patrick did web development. And you know, when you're a kid, getting paid thousands of euros to build a website is like infinite money. I didn't know such large amounts of money. exist, and obviously to a small business. So like, wow, I'm totally taking advantage of this kid from here.

A few thousand euros. And then another common one seems to be dropshipping arbitrage. You know, this whole thing of like buying on AliExpress and selling on eBay or something like that. But you maybe have a window into this with Shopify or something. Do you have a sense of what the right, like if a... 13, 14-year-old is listening to this and interested in making money.

What are the areas where you think are kind of good to pursue today? Yeah, I think these reps are the important thing. The mechanisms change and there's like entrepreneurship clubs are really... using Shopify a lot. I love that. This is one of those areas I would love to go much deeper on, just because it's such a... It would be so fun to really, really build directly for them. The super important thing is just go through...

life like understanding that like everything around you is built you wrote one of my favorite tweets of all time probably like actually maybe a high water mark on X like on this topic like the world's museum of passion projects that's right and I love that because it's like if you can teach that to students it's worth more than most of the entire curriculums of subjects. And so I think the commonality behind entrepreneurs is that they early in life

they somehow something happened that told them that the world is dynamic and it changes. Some of us, I had the enormous advantage of living exactly sort of at the time when events of... computers to no computers like in homes and like I had one early and lots of my neighbors didn't so I had the counterfactual there of and I'm being just absolutely fascinated with them and then networks came in my teens internet

So I had a front row seat to a world, like the world around me in this small sleepy town in Germany was irreconcilable with what was becoming possible. Like it was so clear that so much would change. I think that dissuade me of sleepwalking through the world. It just made me like this, like, okay, like, that's fun.

Like I'm sitting on a piece of information that everyone around me doesn't believe yet, but it's clearly true and I'm going to bet on. And so I'm going to take reps looking at things around me and figure out how they could be done better and like that. process is, I think, the most valuable process to cultivate as a young adult or teenager. So the meta process is more important than the actual...

I agree. Yeah, absolutely. I think if you spot opportunities, see what you can do with it. It's a really important thing. I think we overemphasize both stories and de-emphasize. the process that causes this. It's building the habit of looking at things and thinking about, A, do I know how to make a better version of this, which you should do anyway? And then...

We definitely underestimate and under-celebrate and under-lionize. The Beatles Hamburg days. And acting on this impulse would mean I'm going to spend the next five years of my life doing this. I'm willing to pay what that costs. And for you, is that just doing a lot of programming as a teenager? Programming is how I just made things different.

I went to this classical language school, so I was bad at Latin. We started with Latin. I only got English in like an eighth grade, hence my accent. But your Latin accent is excellent. Yeah, I imagine. Well, no, because no one knows how that actually sounds, and it's definitely not... German. Anyway, I just hated doing the vocabulary tests. I was bad at it, so at some point I wrote software for teaching myself, basically.

similar to i suppose what's called spaced repetition now but like just like and it's like there's a little basic shell script right like and it's stupid but like something about making this and then using it made it interesting enough that I did it. And so it's a very clear moment where I went from, I think I got a D on a test, which was starting to be problematic. I was just like, A's for the rest of my life. That is a crazy experience. This is a thing that didn't exist, that fits.

me personally, maybe my first internal tool. So, yeah, it changes you. Exactly. It made you passionate about building your own software. You mentioned spaced repetition. The other... Interesting, you know, educational topic du jour is mastery learning in the Bloom Two Sigma effect. How do you put that into practice, do you, with your own kids? Yeah, so my kids. I have three boys, 15, 13, 11. And they go to a school, a boys school, which we are quite fond of.

I mean, even my kids, they're quite nerdy. I've met them, yeah. So I love them to bits. I can't wait to go home. It's such a fun time. Everyone should have kids. And the particular mechanisms by... I don't know. It's like the most important principle of like parity principle I have is everything's interesting. This is a sentence you will hear a lot in our household. It's the kids say it.

no one is allowed to say they're bad at something they're only allowed ever allowed to say they are not good at something yet so that's basically the entire set of principles we're not good at gross margins yet exactly yes we choose not yes um you know and making a joke is funny too right like it's just like it's I want them to have those things in their priors because it's really easy to say something is annoying, but it's much, much harder to figure out.

what you would have to do to make it interesting for yourself. And I think there just isn't anything that is not actually interesting. You spend a long time in banking. It doesn't sound terribly interesting to me initially, but I've learned enough about it. I've read books about double-entry accounting. and the history thereof. And it's just absolutely fascinating. Did we recommend the Calamira's book to you?

Fragile by Design, which is a history of various countries' banking system. I'm literally just reading it. Oh, you are? Okay. The Canada chapter is particularly good. It's like a history of the Canadian banking system. Like, very fond of Canada right now, which is like, it's a new origin, I think.

I found it interesting how it kind of explains the path dependence of how each country gets to where it is. Speaking of Canada, what's your advice for Mark Carney? Yeah, I mean, it's like funny. Our partnership is very, very storied and layered and somehow your board members know my...

Prime Minister. So that's fun. Look, I love Canada. I live more in Canada than Germany for longer. I think it's a... It's a beautiful project. It's a beautiful implementation of different implementation from the United States or from Europe of classical liberalism. It's very much veered into socialism under Trudeau. He pushed that way, way hard.

I am very much hoping that Mark is helping all of Canada to figure out that that's a successful discovery of something that doesn't work because obviously... that kind of push socialism always leads to poverty. The problem with socializing is eventually run out of other people's money. Exactly. And so it's just like, and it doesn't need it because it's such an incredibly...

CARE's largest company now at this point by market cap. As an immigrant, I think the advantage I had as an immigrant is I wasn't sort of aware that we are supposed to, you know, make small plans so I made big plans you know just like and held everyone accountable to that and it turns out you can build world-class companies if you do that and there doesn't need to be any kind of inferiority like there and also just like on

first principles is like it's just like i mean first of all it's it is attached to like next door to the best consumer market on planet earth until 15 minutes ago it was we were good friends and um it's uh um you know it it has in the ground basically as much like it's number one two or three on

every single resource that is of value in the next manufacturing buildup age. So like we should probably get on extracting some of those resources and do something. I think it can be richest country in the world if it shows so. and it just needs to make a decision that it wants that. And I'm very optimistic about Marc, you know, pushing into that direction. And I hope he's going to push very far because I think that's...

There's no advantage of waiting. So taking economic growth and GDP growth seriously, which has obviously plateaued somewhat in Canada in recent years, and then being willing to make the potentially controversial... policy decisions and kind of changes that are required to actually enable that? Well, I mean, the things that are controversial policy decisions are just, like, controversial for path dependence, not for real. Like, it's like Canada has something crazy to do.

tell people but that has like this incredible heartburn over over pipelines which are like things that go underground next to highways and just like it's just like no one knows that there's a pipe like the only thing a pipeline does is you don't need diesel trucks that's the only thing and you can save a lot of diesel

It also changes the cost basis and Canada has all these resources and they need to somehow get to ports. And so it just feels like these tiny little implementation details. It's like as if computer manufacturers would somehow get down on copper cables. It's a weird thing. And so deep politicizing, just like implementation details that professionals can do. It's odd because Canada, for instance, doesn't have its own nuclear power. Like where I live in Ontario, it's like...

the grid is entirely clean. It's like the Bruce nuclear power plant, which we constantly build more reactors for, which is like every time under budget and before schedule, like six, seven gigawatts now. just humming and then the rest is it's the cleanest grid in the world yeah and so like the funny thing is it doesn't even need it because it has all the other stuff too that it could just use but like we could have this too so like i think there's like

the blueprint is obvious. But actually, so my main advice to him is always like, I want him to, in fact, ideally talk to you about the Shannon Economic Zone because I feel like it's like a blueprint for success in Canada because I think we can build data centers or space ports and stuff like this much faster.

My hope is that Canada can build up an economy and a meritful market and so on, which makes it like... can get back to doing what Canada does best which is like succeed by helping America succeed and you know just like work together and let's go.

And as you say, energy could be a big part of that. You mentioned nuclear. I was thinking about hydro, where obviously Canada has a ton of hydro already. Like you were saying, the seven gigawatt nuclear power station. But I think La Grande Rivière is a seven gigawatt project and could be a lot more.

Other phases that were not done. Absolutely massive. On every single angle you look, it is every single time I learn about a new resource that we will probably need a lot of. If I look, it's like just like there. Yeah. Like it's just unexplored. Yeah. Maybe last question. We have your Daytona car here. What is the draw of motor racing to you? Yeah. I mean...

I just love it, and it's surprising to me. Honestly, this is one of the... Sorry, is it or isn't surprising? It is surprising to me. So it's like... I love kind of figuring out my own limitations or propensities. I think it's a game everyone is playing, I think, to some degree, maybe less conscientiously. try on every hobby that anyone I admire gets into because it's like, maybe I'm interested in it too. I wouldn't know. Again, everything's interesting.

Motorsports was like not really something I high likelihood or betting average on liking because I didn't even have a driver license after coming to Canada. after my German one expired. So a couple of friends went to a track and I went out and just immediately fell in love with it. There's something just incredible and I love competing against myself, I guess.

i think that's that's the thing i i just like i race in races and um i don't even try to win i'm like i'm i'm still competing against myself and i see all the other cars as obstacles um and um so there's just like this beautiful thing about it's technical and you connect it with this like four basically playing card size patches of rubber to the ground and there's only so much grip. to get out of them and like trying to get to like 99.9 percent of like

lateral acceleration or turning, understanding the weight transitions of the cars and all these kind of things. And trying to get to the best lap time you can. And the difference between that lap time and the theoretical possible lap time is just that...

you know, exact report card of all your inadequacies as a driver. And like next lap, you try to do it slightly better and better and better and just hone this craft. It's like, just like the immediacy and the adrenaline. It's just like, it's a... It's just a brilliant thing which I discovered for myself. So I started taking it more seriously over a couple of years and now like...

the toner this year and hopefully do the big 24-hour. Which is a 24-hour relay race, right? Like you swap with other drivers? Yeah, that's right. So one car, 24. It's actually, if you watch the Formula 1 cinema movie, that starts.

at that race which was really fun to see it's like i've done the getting up in the middle of the night walk exactly down that uh first scene the bread pit walks like down to the pits and um stall was even exactly the one you went to wear like in a movie and then get in the car and you're very very tired in the middle of night so i have to say i found out recently they sold sponsorships for the uh

fake F1 teams in that movie and made a lot of money on that. Incredible bit of product placement. If you think about it, Formula One itself is such a made-up thing. It's so synthetic. We just decided to... create a document, which is a set of rules. And somehow, because that document exists, like 20 teams move cars around the planet, which is also one of the bull cases. I don't even believe in...

AI unemployment to begin with is not how things work. But if you were worried about it, F1 is a good... F1 is kind of proving that we can just make up completely incredible pursuits, especially with... AI and robotics, we can make it cheaper and then like, let's go. Yeah, there's a great, I mean, the history of F1 is super interesting. There was a great Colossus podcast on the history of F1 describing basically the Bernie Ecclestone era, which came with, you know, the emergence of TV, obviously.

And there's even things like the different race schedules were not standardized and there was like multiple, like not every team would show up to every race or whatever. And so he really kind of... harangued everyone into kind of a standardized thing. But then it feels like the modern era, especially with Netflix, I think they've very successfully internalized that they're selling an entertainment product. And some people will...

watch every race every weekend, but I'm not going to watch every F1 race, but I will watch Drive to Survive. That is an appropriate short-form amount of F1 for me. And so I think they've actually very savvily internalized that they're... F1 is an entertainment product and it needs to be kind of consumable in all the appropriate ways.

And they just gave access to the sort of intrigue and politics and just characters behind it. Formula One drives up crazy people, right? It's a nutty thing to basically drive a billboard around at 350 km an hour. So like the characters are... larger than life and absolutely driven by almost zero-sum competitors. No one's there to just compete against themselves. And part of how it works, I remember hearing Zac Brown, the McLaren team principal, talk about how

Netflix is everywhere in the paddocks and around and things like that. And he was saying, yes, you're aware that you've signed a thing that Netflix microphones will be everywhere, but they're really savvy where they'll have the, like in a spy movie, they'll have the directional microphones trained on you from like a super... far away distance and you just end up forgetting that like you're on a hot mic your entire day while trying to actually manage a race and so I think they get

authentic content because yeah in theory people know that they're on camera and on a microphone all the time but you actually just can't live your life that way you know you just forget and so they actually get kind of interesting content i completely agree and it's like the season is long they do many It's actually amazing what they do. They're fantastic athletes. They lose five.

10 pounds. Yeah, yeah. Singapore night race. So it's pretty amazing and it's a fantastic sport. I actually love what Netflix did because Liberty Media is one of those companies you kind of have to start. Termalone. Cable Cowboys.

It's already one of the best. It's a great book. It's such a great book. You know he wrote a book. It's coming out soon, imminently. I did not know it. Yeah, he has an autobiography. That'll be firebooks. Yeah, it'll be really good. I agree. Well, look, thank you, Toby. This was fun. This was super fun. So it's really, really fortuitous when an Irish pub just popped up in your office.