Great. Thank you everybody for joining us. With us, we have the CFO of Shopify, Jeff Hoffmeister. Jeff, thank you so much for taking the time to come to London and get in front of these investors. Of course. Appreciate it. Let's kick off on something that we spent a lot of time on the breakout meetings on, it dominates the earnings calls. I think it's dominating the broader technology zeitgeist as well. If I just zoom out, regardless of Shopify, what is the lay of the land of the agentic commerce landscape today from your perspective? Yeah. I think part of the issue with agentic commerce is everyone has their own definition of exactly what this looks like and when we get there. I want to make sure we differentiate agentic commerce between versus what is now what I'll call LLM-driven discovery. Right? We see a lot of activity on Google Gemini, on OpenAI, and a lot of LLM platforms where consumers are starting there to do their discovery of a potential purchase and discovering a merchant. As it relates to agentic commerce, I think that's a vision for most people feels a little bit like there's a point in time where I'm giving my LLM a budget, or it's proactively deciding for me what to buy and how and when do we get there. We're obviously still early stages in that. It's one of the things that we, as you know, we partnered with Google to build UCP. I think that's one of the key things for us, which is really helping make that future possible sooner rather than later in terms of all the rails that make agentic commerce happen. The adoption of agentic commerce will of course take a bunch of different forms, and it's probably going to be heavily dependent on certain demographics and other things, and just there will be some segments which are early adopters and some which are slower-paced adopters. There are some people who, there's a lot of different views on how quickly we get there and what form of it we get to. Certainly from a Shopify perspective, we do believe in agentic commerce future. We think it's going to happen. We're still in the early days of all this. There's a lot I want to unpack there. Let's start first with, I think there's confusion. People hear Universal Commerce Protocol, they think of Google, and they tend to mention Shopify second. I'm going to let you clarify. Talk to us about how UCP came about and what the different roles the two companies played in establishing the Universal Commerce Protocol. We've been a partner of Google's been a partner of ours for a long time, and this was really a co-development effort in terms of, you can imagine from a Google perspective, they have, of course, tons of experience, pattern recognition, thoughts on how you think about the discovery, the front end, and we of course, have done more historically on the commerce side. I don't think there's a bifurcation of this party did this versus this party did that, but this is something where we both worked together to build something which we thought would broadly, as I alluded to before, really enable agentic commerce in a way which it had not, to that point, been manifested. We both had a team of engineers that brought their expertise to bear. I think it has manifested itself in something which is really going to help usher in the next stage of this, and you can see, just based on the adoption of a bunch of others, both retailers and other platforms that have followed behind us and have done the adoption of UCP. I think you can see there's broad industry support from a lot of people that this is heading in the right direction. When you think about the, you've mentioned rails a couple of times, what does UCP enable today for the merchant customer of Shopify's? When they're thinking about embracing it, what steps do they have to take to enable the utilization of that? Yeah. No, it's very easy. Again, UCP is the protocol that helps make all this happen, and the manifestation as it relates to Shopify merchants usually starts with something like our Agentic plan or with Catalog. Right? Catalog is something that we set in motion more than two years ago. Think of it to a certain degree, like a large taxonomy of all the merchant's products. We have over a billion products in there now. It really helps you as a merchant make sure that your products are discovered, if the consumer's starting on one of the LLMs, that your products are discovered in a way that you intend, that it's not a screen scrape. That it's a full encapsulation of everything you've done in terms of the pictures, the product descriptions, the details, all the specifications, everything that you've tried to associate with your brand and spent a lot of time and effort, and dollars, and management attention in capturing your product in the exact way you want. Catalog helps make that happen, and that's one of the things that UCP enables. The Agentic plan is one of the things that builds on top of that to really give, especially the large enterprises, the ability to maybe start with Shopify there, and that's the onboarding of, hopefully at some point adopting the full commerce stack. You kicked off with this about everybody has different definitions for what agentic commerce is. I think today, to the extent there's traffic, it's more LLM search, which you mentioned. How far are we from that reality of, I'll use an example someone else gave rather than one that you gave in the group meetings. Okay. I've heard the, "Here's my grocery shopping list," or, "This is what I want to eat this week. Yeah. Find me all the ingredients, and I want this specific ingredient across three different grocery stores if you have to, but I don't want any substitutions." How far are we from that world where I say, "Here's my $300 from my grocery budget. Go and get me what I want to prepare this meal"? I think there's going to be a difference between traditional discretionary items, things you buy all the time, like food and basic supplies, so to speak, versus how we, and you see this in the Shopify merchant base, how a lot of our brands, we serve a broad swath of merchants. On the margin, they tend to be a little bit more discretionary in some of the things that they do. A lot of this is an adoption of, again, either I have a specific budget I want to give on a monthly basis, or I want to send an agent out, maybe it's time for my kids to go back to school. Maybe I'm going to a trip in a climate which is a different climate than where I live, and so I need warmer clothes for colder weather, whatever it may be. This is very tough to predict. We're probably a year or two, at least for most people, to have that full control that they're ceding to the agent. Part of it is just how you as a consumer want to purchase. Some people want to go into a store, they want to look and feel, they want to feel very comfortable about returns and all that stuff, and other people are just to the spot, and some people just enjoy the shopping experience. They don't want it handed over to an agent. That's part of what they want to do is experience the whole process, for example, maybe discover a brand which is an up-and-coming brand, which is something that they maybe otherwise would not have found. There's a bunch of elements to it, but I think, Samad, we're a couple of years away from that, at least in my early estimation of when we get a meaningful percentage of people that are shopping that way. I'm guessing if it's like anything else, we'll probably overestimate in the short term, underestimate in the long term what the impact. I think that's fair. Probably. I can't wait for somebody to do my grocery shopping for me. I can't wait for that day. While we're on the topic of AI and different relationships, I know Shopify is also working with OpenAI in different contexts. Can you maybe help the audience understand what the checkout inside of ChatGPT is like and what Shopify's role in that is? We've had a partnership with OpenAI in various forms for more than a couple of years, and some of that started with how we use them internally, and obviously, the partnership has evolved in terms of some of the things they're doing in agentic commerce and LLM-driven discovery. From our vantage, if you're doing a search with OpenAI, you'll be channeled back through Shopify Payments. It's still done on our payments rails. If you're a merchant, you want to have as much of the activity back on your website. So whether it's captured on the OpenAI website or whether it's funneled back to the actual merchant's website, you're still operating with the Shopify platform and all the capabilities and everything that we do, and the economics to us are the same. Whether you start on OpenAI or whether you start on the merchant's website, there's no difference to us. Understood. Just as I think about, we've talked about the agentic commerce experience itself. Let's maybe switch gears and think about AI and what you're doing for the merchant with your different AI solutions, right? I know there's been a lot of focus on Sidekick. Yep. Maybe help us understand what type of merchant is utilizing it today and what the early experiences look like. Yeah. The great thing about Sidekick is it's really gotten to a spot where merchants of all sizes are asking all types of questions. If you'd roll the tape, again, I alluded to this before, we've had it out for more than a couple of years. If you rolled the tape back 9- 12 months ago, it generally was a little bit more some of the medium-sized merchants that were asking questions of it, and they tended to be a little bit more go-to-market-type questions, like maybe I expanded this geography, or I want to expand into Latin America, which specific country, as an example, or how do I think about pricing, or things that are a little bit more of that ilk. Now it's really merchants of all sizes asking all types of questions. You can imagine a merchant that has two employees, five employees, is asking a different type of question than some of the largest multinationals that maybe are asking much more specific questions. They still fall in the full spectrum of everything from how do I think about designing my website versus what do I sell versus how do I price versus all the go-to-market things versus maybe what is the next best product that I should build or sell based on what I'm seeing from competitors or the success. It's literally meant to answer any question that you may have as an entrepreneur. We track it very closely in terms of the types of questions, which merchants are power users, what type of incremental success that they're having. There's, as you know, there gets to be a little bit of correlation versus causation in terms of the data we see for the merchants that are power users. Needless to say, the feedback has been exceptional, and we're really helping it see merchants accelerate their business. Of course, that then plays back to us in terms of additional payments revenues and additional solutions, whether it's FX or tax or other things. It's driving merchant success, which helps us. I'm going to go with causation. This joke that I'm going to tell is only for two people. Because Shopify's stock is up 5% right now, I said if you do stuff with us, it'll be up. There you go. That's not correlation. You're the causation. It's causation. You're the causation. Just like your merchant success is causation. I'm going in that camp as far as I'm concerned. Great. Well, thank Well, thank you. So- I will gladly take it. I want to pull more on Sidekick because I see how it helps the merchant, but if you think about what the biggest opportunity is along different components, do you look at it as something that is a more of a top-of-the-funnel demand driver for merchants? Is it more about improving conversion or retention there? Is it new monetization? Without stack ranking them, how do you think about maybe the biggest opportunities along those different vectors? Yeah. Unfortunately, it's a little bit of everything. It truly is merchant success. There's some interesting things that we've seen in terms of types of merchants that are benefiting, some versus others, but they're all using it in their own way. I wish I could tell you it's more conversion, or it's more this, it's more that. I think for the average merchant, again, it's still mostly go-to-market-type questions that they're asking of it, but it continues to evolve in a way that I, again, like its intention. Ask any question you want, and let's give you the best answer we can. We've talked about what it unlocks for a merchant, but an investor question that's come up, and you mentioned on the last earnings call, is maybe the costs associated with it. Yeah. I think what everybody in the investment community is trying to figure out is. A lot of companies are giving AI tools or enabling it, and they're absorbing the cost up until now. What's Shopify's philosophy? What have you experienced so far? I have a couple of follow-ups. Well, I think one of the things, to your point, there's been discussion, a bunch of questions on it. I think one of the things that's lost in a lot of these conversations is the impact that it's having on revenue. I think a lot of people look at this and say, "All right, how do I think about the cost?" That's true. Of course, if it drives much more merchant success, then it continues to sustain these growth rates, which we've been delivering, right? Four straight quarters of 30% or more revenue growth. You see the free cash flow margins doing what they've been doing. All that success is, I'm of course not going to attribute that only to Sidekick. It's a bundle of all the things that we do. To the extent that Sidekick can really help get more merchants on-platform, help them stay on the platform longer, help them increase their conversion. There's all kinds of statistics we see about, from their vantage point, what are they doing to be more successful? As that drives the revenue, then of course from an LLM cost perspective, we monitor that very actively. There's a bunch of different things we do internally to track that on a daily and a weekly basis. One of the beauties of Sidekick is it's not the value there is 20 years of commerce data built into Sidekick. It's data that we and only we have because it's all the activity, the before the purchase, the at the time of purchase, the after purchase, all the context, all these things. It's not, hey, I sent you to a website and now you went and did something on a merchant's website. It's we run the merchant's website, and we see all this information that's in there. What we don't need to do is have the latest and greatest and most expensive LLM run through all that data in terms of both inference and training, because really what we have is the ability to take a model that maybe is not the latest and greatest, or maybe it's an open-source model, and how do we think about triaging, stacking, implementing the right model for the right point in time. We want to make sure that when we have a question from a merchant, we're delivering the best answer possible. The other thing too is you can imagine a lot of the questions can be somewhat similar in nature. I said smaller merchants will maybe ask different types of questions than larger merchants. There's an element though that it's pretty similar across all those. How can we do things, for example, like caching of answers or anticipation of questions, which also allows us to keep the cost down. As you know, because Tobi's memo came out more than a year ago, we've been a thought leader on how you think about uses of AI. You can assume that we extend that knowledge, those best practices into how we manage the cost and how we think about it. I'll quote something that you've mentioned in other meetings. I think it's worth noting that Sam Altman himself said that Tobi's using AI the best, or that Shopify is using AI the best currently, and I think that's a testament to what you guys have done there. Maybe before we wrap out and switch gears, you mentioned data as one of the big moats for, or one of the value centers for Sidekick, and then I think a moat more broadly speaking. What would you describe as the other key moats in a world where everybody's trying to figure out what's at risk? I think checkout is something that's misunderstood. Just help us understand maybe the moats that Shopify has. Yeah, the data piece is one of it, but the data, of course, then extends into a manifestation of the pattern recognition and all the things that help us think about the next product to build. You see, again, Catalog and Sidekick are things that we set in motion a long time ago that no one else has even really started to build. You need to have the data to build it, you need to have the product expertise to go ahead and actually deliver on the products that we have. I think, as you also alluded to, Samana, the checkout piece I think is one of the things that is most misunderstood or least appreciated in terms of how hard it is to do because you really have to marry the product expertise and the speed of building something that operates in milliseconds in a lot of parts in the process. When you see the statistics in terms of the actual impact on abandoned cart and lost dollars for a merchant when a checkout process is either a little clunky or not super fast or doesn't have all the context on you as a buyer, on the shipping data, on the return, all that stuff matters in terms of the success rate of you going and completing the purchase. There's a lot of things that not only one, you have to start with the ability to build that from a speed perspective and have the technology expertise to do that, which some others would do, but what they don't have then is all the context around that, which one, makes the checkout process better, and two, allows you to build an even faster checkout process. Again, as it relates to the tax and the shipping and or FX, like all these things that others have tried and attempted to do and they've not been successful doing. I think that's the piece that's I think sometimes hardest to describe but also most underappreciated. AI takes up a lot of the oxygen in almost every conversation these days. Yeah. I think Shopify has a lot of different growth opportunities that are doing quite well, and I'd probably argue up until this point, those have all been doing much better than AI itself, right? It's very early. Let's unpack some of the other success categories that you have. Maybe apropos of the audience, international has been- Yeah. ... a very high-growth area. Can you help us understand what's changed on the international front? Is it the go-to-market has either changed? Is it that the product market fit is there? What has happened that's really driven that international success? Yeah. Well, as you know, the biggest international success for us has been here in Europe, outside of North America, of course, our home markets. It really a couple of years ago started to take off, and I think that was mostly a function of product market fit. That product market fit as it relates to some of the things we're doing on kind of the core product offering then begets additional adoption of other things. We roll out payments in more countries, merchants start taking on more elements of sale. They take on some of the cross-border stuff we do. They take on some of the tax stuff we do, especially given some of the tax dynamics here in Europe, et cetera. That then begets us in a position to roll out more products in more geographies. We really start very much from a product perspective, the marketing dollars flow behind that. Yes, once you have the product market fit and increased merchant adoption, you put some marketing dollars behind it. I mentioned not on this most recent earnings call, but the one before that roughly 40% of our performance marketing last year had been filtered into Europe, and that really helps accelerate some of the things we're seeing. As you know, last year we rolled out payments in 15 more countries in Europe. We continued to roll out the full integrated kind of latest version of our point of sale. We continued to roll out capital in more regions. We've really gotten to the point where this is a very repeatable process. I think we have payments now in 40 countries, essentially Europe fully covered. There's more to do this year and next year as well in terms of doing more in Europe. The opportunity for us is to really take that same playbook. There will be some modifications of it, but localization with AI has gotten as easy as it's ever been, right? Localization of the software, localization of the advertising, localization of everything the merchant interacts with. Take some of the things we've done in Europe, one, continue the momentum in Europe, two, think about, well, how do you do that same playbook in Latin America? Where you look at Latin America, we have merchants in all the countries in Latin America. We have some great merchants down there. But as a percentage of our revenue, it's still a small single-digit percentage. Same in Middle East, same with Southeast Asia, same with Africa. There's all these regions where you say, take the European playbook, replicate it in some way, shape, or form, and continue to roll this out. Even without the introduction of any new products, and of course, we will continue to do that, you can roll this out in a way which I think gives us durable growth for a long time. I want to maybe unpack that a little bit on one of the big drivers for you or for Shopify overall has been driving the merchant take rate higher, the overall take rate higher. Yep. How do you think about the progression for merchants outside of the United States in attach for things like payments, capital? I know you mentioned that you've deployed it now into several more countries. Yep. Should the long arc of that look fairly similar between the U.S. and other geos, or would it look different, and if so, why? No, it looks not identical, but very similar. We've seen very few differences, not only in attach rate, but just in general, as you think about it, not surprisingly, commerce is pretty straightforward. You, as a consumer, are trying to find a merchant. There's an exchange of money, the product is shipped, et cetera. The core engine, the core process, the core guts of commerce are very similar. As you think about what a merchant and what a brand is trying to do as they grow over time, there's a natural progression. As they get bigger, they're more likely to take on more products. They're more likely to do different things. They, for example, are more likely to get into cross-border. Yes, on day one, you can do cross-border things. As you get bigger, you're more likely to do that. You're more likely to do some things on the tax side. The opportunity for us, and you look at the attach, we don't break out attach rate by geography, but if you look at the attach rate in North America versus the attach rate in Europe, there's clearly opportunity. A little bit like payments. When you look at payments penetration that we have in North America versus what we have in Europe, there's clearly headroom there, too. We also see this in terms of cohorts. As you know, we think about our business a lot in terms of cohorts and how do you look at one cohort by year or one cohort by geography, and how do those stack over time? Very similar in terms of their trajectories, their growth rates. The opportunity for us is to give the merchants in Europe the same set of products that we have in North America and et cetera, et cetera, in terms of roll it out. Understood. You mentioned payments multiple times there, and the penetration's very high, especially with U.S. merchants. How much more runway is there? Is it purely just rolling it out into additional geographies, or are there other gating factors that have kept payments penetration maybe outside of the U.S. looking different, and how should the shape of that look going forward? Yeah. Similar to the comments we just discussed, there's no difference, it's just a matter of continuing that progression. We have not broken out payments by region, but as you would suspect, the payments penetration in North America is higher than it is in Europe. With the additional country, the 15 countries we rolled out last year, part of this is getting those countries' payment penetration levels higher in those, continuing to do local payment methods, which is something which, especially in Europe and as we do more internationally, helps on the capture rate. You get a merchant on platform and a higher percentage of their total transaction activity happens on Shopify. Of course, there are some things like cash, it would seem like there's less and less of that. There'll be a natural governor to avoid us getting to 100% penetration rate. We still see a lot of opportunity in North America, and our payments penetration rate in Q1 was up 3 points over Q1 in 2026 versus Q1 in 2025. There'll be some natural headwinds in Europe for these countries where we just rolled out payments in terms of the total payments penetration level. That should turn into a tailwind as these countries continue to have more success and that rolls out further and further. I want to talk about Shop Pay specifically. Yeah. The checkout function, which is, it's dangerous for an impulse shopper like myself because you don't have to enter in anything. You just basically give your number. Yes. It's trouble, is the best way I'll put it. Help us understand what the value proposition for the merchant there is. Yep. Why that drives conversion, which is probably the trickiest thing for a merchant, I'm guessing. It's exactly what you just alluded to. Because if a merchant knows that if they offer Shop Pay, that it's going to help on their conversion, that consumers will go there, they will see a brand which they trust for good reason. Shop Pay has the brand and the trust and the awareness that it has, not because we spent a bunch of marketing dollars on it. It's because we built a product that is exceptional in terms of, again, all the context and the speed of everything it does, which has created that level of trust that you and other consumers have. Shop Pay, along with the Shop App, of course, are the two things where we have a consumer-facing element to our business. That is a consumer-facing brand, which is there to lend strength to what we're trying to do on the merchant side. If you go to a merchant and you can say, all right, think about the impact of Shop Pay on your abandoned cart. Literally take what your abandoned cart number was last year, multiply it by the uplift in conversion that you get from Shop Pay, and basically explain to me why it is that you want to let those dollars just leak out the door, so to speak. It's a very powerful conversation to have with them. It has clear, tangible impact on their business, and that's a function of, Tobi likes to use the word flywheel a lot. It's a function of that flywheel, right? It's the consumer brand which lends to the merchant, which allows the merchant to be more successful, and it continues in that pace. I believe a few years ago, Shopify made Shop Pay available for non-Shopify merchants. Right? I think it's clear how it expands the opportunity, how does the adoption of that look like, or how has the reception been, and has that led to merchants actually wanting to adopt the core platform as well based on that experience? Yeah. It has. We've been very careful on how widely we do that. We are starting to do it more. The Agentic plan is also part of that initiative in terms of how do we look at the evolution of payments and Shop Pay, and where does this take us? I don't have statistics to share with you in terms of what percentage of Shop Pay standalone has then led to merchants taking the full platform. I would say across the board, everything we've been doing in product development over the last year or two has really put us in a spot where, I think from a competitive dynamic, we're doing some pretty incredible things. Let me take a step back. Some merchants will be at a spot where they'll say, "I want only so many wallets on my checkout," and others are like, "The more the merrier." For the merchants who get to the point, "Well, hey, I have three, four, or five, whatever checkout methods on my checkout page. Do I want to add another?" It's very hard to refute the statistics of Shop Pay and what that does for the business and the conversion and the uplift and everything we've talked about. It's just a matter of kind of getting them to add that, and that kind of leads to obviously a lot more in terms of the upsell. Great. I know when most people think about Shopify, they think online for obvious reasons. The in-store or the retail side of it has gotten really large in brick and mortar as well. How do you think about the point of sale opportunity from here? How do you get more of your merchants that have had tremendous success online to put it inside the store as well? Yeah. The thing that we offer in our point of sale solution, which really only we offer, is the best technology on both the online commerce stack, which of course we've had for a long time, married with the best technology on the offline side, and doing it in a way where you have one platform, set of analytics, everything to look at in terms of your customer information, your inventory information, all your transaction, all your payments, all that stuff in one platform. That's been one of the pulls for us on the point-of-sale solution. Roughly 80% of commerce, as you know, is still offline, that's something that we want to continue to grow, and we will continue to grow, and the traction has been there. The other thing that we've seen really starting about a year ago is up until then, most of the stuff we were doing on point of sale would've started with an online merchant. Now we're seeing a lot of, and Harley's talked about some of these on some of the earnings calls in terms of some of the brands that have started on the point of sale or have taken us for the point of sale only solution, and what that's meant. We're getting wins on both sides of the equation. It also helps, especially on the large enterprises, as we think about going to them and just basically saying, "All right, so we have point of sale." We have point of sale, we have B2B, we have all the cross-border stuff, we have all the tax stuff. We have all these things where it becomes hard for them to say that we don't have everything they need, especially with the addition of all the Agentic plan. You mentioned large enterprise just now. You guys have had a lot of success with several large logos. I think a notable one is LVMH recently announced. Help us understand what has changed on the enterprise side that's bringing more of those larger merchants that are maybe a little bit more hesitant to embrace change. What's driving them to you now? I think it's a couple of things. Part of a multi-year effort too. As you know, it's been roughly four years ago when we put a particularly concentrated effort into getting more and more merchants. It was never a technology question because we, even before then, were doing some of the biggest flash sales on the planet, and so we could handle all the activity. It was just a matter of how do you get a go-to-market motion, which gets that technology, that product capability out in front of the largest enterprises. As you'd naturally start with, maybe we'll call them the small end of the large enterprises, or call them medium-sized merchants if you want, and then you get those wins, which begets slightly larger enterprises, and continues to build on itself in terms of size. We've seen a lot of that. We've also seen an interesting dynamic, too, where a lot of times some of the largest merchants look at what some of the small and up-and-coming brands are doing in terms of everything from kind of the customization and kind of the presentation of the brand and all the interesting things they're doing on the commerce side, and they look at some of the smaller brands and say, "I should be using that platform." You see some of the biggest fashion houses, for example, look at, again, some of the emerging brands and be like, "I should be doing that." It's interesting because you would've thought some of the small ones would look to some of the big ones rather than the reverse. We actually see both of those dynamics at play. Part of it is that momentum. The other thing, too, is just from a technology and a product perspective. That's all the things we just talked about. Point of sale has continued to grow and continued to strengthen in terms of what that solution offers. Our cross-border product, our Managed Markets product continues to get better. The stuff that we're doing in agentic, no one else has anything like that. If you are a large enterprise and you're looking to move your platform, whether you're moving off an existing solution from another provider or you're moving off an in-house solution, how do you justify moving to a platform that doesn't have an Irrespective of when you think agentic plays out, how quickly it plays out, to what magnitude it plays out, do you really want to adopt a platform that doesn't have a thoughtful, strong, proven offering in agentic? That's hard to justify. What we're doing in terms of differentiating on that side has helped a lot, too. Part of it is, again, as you know, is just the length of the sales cycle. Even with everything we're doing, the sales cycle generally can be 9, 12, 18 months long, and the bigger the enterprise, generally the longer it takes. The more products that they take, sometimes that can slow it down because you may say, or they may say to us, "We're completely good on the commerce stack. I have a couple questions on point of sale, or I have a couple questions on this or that," and kind of it just elongates it. The win rate has been very strong. If you think about, you mentioned AI, and you'd want to buy the future-proofed solution, right? Yep. Or you want to bet on the technologies of the organization that's laser-focused on that. Has it changed actually where maybe the starting point in the conversation is with the large enterprise, where you may have originally talked to, let's call it the head of e-commerce? Is it now you're talking to the board or the CEO? Has it changed either who the stakeholders are and/or where the conversation starts, and does that tilt things in your favor more? It's changed a little bit, but it actually has not changed as much as you would think, because especially if you're displacing an in-house solution, there's a natural desire, if you spent a lot of time, effort, years, management attention on building something and customizing it exactly the way you want, there's going to be, for a host of reasons, just starting with human nature, there's going to be some resistance about it. That's something that we've faced for a few years. That's still there. Yes, the stronger our solutions get, the more there is a pull to having conversations with the CEO, with the founder. The selling motion actually has not changed that much, only because even a couple of years ago, you're still dealing with a lot of the same dynamics. Maybe last on the go-to-market side, in terms of the sales organization, is headcount there where you want it to be? I know there was maybe some compensation changes that you had announced, or that were disclosed late last year. How has that maybe affected change of behavior inside of the sales organization? Two parts to your question. First, from a headcount perspective, yeah, we are where we believe we need to be, and again, it's kind of three or four years ago we really started building out the sales force. You may have changes in the composition of the team, but you still, from the size of the sales force, we're still doing what we need to do, and as you know, headcount has come down each of the last three years. It's kind of stepped down slowly each year. That's something we've been very disciplined about. As it relates to some of the changes in the compensation system, as you know, big enterprises that sell, big companies that sell into enterprises, you're going to make changes in the compensation plan. You're going to make tweaks. You're going to continue to evolve it. In any given change, some people will like the change more than others or vice versa. That's one of the reasons why in the last earnings call, Harley had a statistic, and I had a statistic as well, both as it relates to merchants. Everyone has their own definition of enterprise. Harley and I talked about GMV of $100 million and above and some of the progress and the strength they're seeing there. That strength and momentum speaks to us doing some good things, and as I alluded to just a couple of minutes ago, the product capabilities gap that we have versus others continues to widen. We feel good about what's going on in the quote enterprise. Look, you're one of the few companies, I believe in all of software right now, that has reduced headcount and accelerated growth. Maybe there's some magic to that as well. It's fairly remarkable. I want to switch gears. We've talked about external AI and the AI e-commerce infrastructure. We've talked about products you're developing. Let's talk about the company's internal innovation, especially as we just talked about headcount. How is Shopify using AI internally, and where have you seen the biggest impact from internal adoption? Well, as you know, a lot of attention on Shopify started with Tobi's memo a little bit over a year ago in terms of how we think about internal uses of AI, and not surprisingly, I think for a lot of people in tech, it starts with R&D product development. We had a lot with support as well in terms of some of the things that we were doing interestingly on the support side. R&D for sure first, support is a fast follower, but it's really even six, nine months ago, it's all parts of the organization. It's everyone from finance to talent/HR to legal to sales. All elements of the organization are using AI in a very thoughtful way. Just like we've been thought leaders in terms of encouraging people to use it, we've also been thought leaders in how do we make sure that we're getting the right model for the right person and the right use case at the right time, such that we just don't have people running off and using the most expensive model when they don't need to. We have circuit breakers that we look at. There's a couple of reports and dashboards in particular our CTO, Mikhail, and I look at, how do we look at those on a weekly basis and find things over week-over-week change? We look at the types of models that are being used by department, by level of seniority, all these things, and kind of when should we be using an open source model versus when should we be using some of the more expensive latest versions of the Anthropic's and the ChatGPT's and Gemini's, et cetera. We're very intentional about that. Part of the reason we can be so intentional about it is because Tobi has such a deep understanding around what model is necessary. As you know, you probably hear as investors, a lot of CFOs talking about, well, how do I manage these costs? From my vantage point, Tobi leads the charge on this stuff. Mikhail and I make sure we're being really thoughtful on this, but Tobi's out there making sure we're being highly intentional. As a follow-up to that, I think people think about the value created by AI internally. They express it in different ways. It's either, hey, it's driving productivity- Yeah. Which maybe drives revenue, versus it makes people more productive, so maybe I can lower headcount or lower costs. How are you thinking philosophically as the CFO between that trade-off of if I get X productivity, I want to maintain headcount, or would I rather take that productivity and reduce headcount? How do you think about essentially a form of reinvesting savings or productivity gains? Yeah. Shopify is, as you've alluded to it, we are a high-growth builder company. Will we, like everyone, be thoughtful around headcount and make sure we're doing the most we can with what we have? Yes. On the margin, we're going to say, "Let's take the engineers we have. Let's continue to build some amazing things." We do Editions twice a year, as you know. We continue to crank out more products. Let's think about taking this team and have them do even more for merchants. That's really the starting philosophy for us. As you think about with headcount, again, for the last three years, it's been essentially down some, you can call it flat if you want, but it's been down, and the growth has continued. Again, four quarters of 30%, three years now, we're going on 25% each quarter. The growth is there. The headcount is flat to down. There's some things, and we've talked a bunch about kind of all the different dynamics and operating expenses, but that just leads you to being able to create more and more opportunities, in terms of what that means for free cash flow margins. When you think about the impact that adopting these tools had on your R&D flywheel. Yep. Can you speak a little bit to, I think that's an area where a lot of software companies have talked about a positive impact. How are you guys thinking about development, and how has that impacted the product shipping that you guys have done? Yeah. Well, we definitely spend a lot of time thinking about for the pull request and kind of the code development, how much is being written by AI versus how much is being written by the engineers, how much testing of code are we doing, and what percentage of the code we're cranking out is being monitored by the LLM. We haven't talked extensively about those percentages, but you can, again, you could assume, given some of the leadership and thoughtfulness we've had and the things we've talked about, that we feel pretty good about where we are and kind of how we're continuing to accelerate this pace. I don't see any slowdown. Again, we have an Editions session coming up, and as we continue to talk about all the things we're cranking out, I don't see any slowdown in any of that. I want to talk about pricing in that context, right? You've rolled out Catalog, you've rolled out Sidekick, you're rolling out, those are some of the headline features, but you're constantly rolling out new product and shipping it to your customers, and that flywheel's accelerating. It's been a few years now since you've raised price on both the standard and the Pro SKUs. How do you think about how adoption of some of the newer features you've rolled out, how will that impact your thoughts on pricing, broadly speaking? Yeah. Well, as you know, we've always thought on kind of the spectrum in terms of value versus maximizing price for merchants. We've always been very focused on the value side of that equation. When we changed the pricing on standard plans three years ago, it had been 13 years, actually, since we changed it. When we changed on + 2 years ago, it had been almost seven years. I think you'll see probably more frequency from us in terms of how frequently we change those prices. Of course, I don't have anything as it relates to pricing to announce today. We'd roll that out with merchants. But from our vantage point, we want to, again, stay focused on value. Sidekick has really helped delivering more value to merchants. Let's help them be more and more successful, that will kind of inure, again, back to our revenue, back to our margins and everything we're trying to do. I think we're very thoughtful in monetization. We're always looking and testing the prices, and we may test, for example, in a specific geography, maybe we're testing the pricing on point of sale, or we're testing the pricing on this plan or various things. Yeah, we're always out there testing, but as it relates to any big pricing plan, I have nothing to announce today. Great. Then, margins is an area where, since you joined, Shopify has seen essentially a step -function increase on GAAP margins and free cash flow margin. I think investors have been really appreciative of the progress there. This year, free cash flow margins, I think, are expected to be largely flattish, plus or minus. Just help us think through maybe the broader free cash flow margin expansion trajectory and what the puts and takes there are. Yeah. I think, if you look at free cash flow margins in 2025 and 2024, you see some pretty significant similarities in terms of two things. Number one, where you ended up for the full year, free cash flow margins, and let's call them roughly high teens for both of those, and then kind of how they progressed through the quarter, right? Generally, kind of Q1 and Q2, had generally been, let's call it kind of mid-teens, and you see a little bit of inflection in Q3, and then Q4 is generally, of course, the big holiday shopping season. While the numbers differ a little bit, that pattern, kind of that intra-quarter pattern throughout the year, again, holds true for both of those years. We're obviously not going to give guidance for the full year this year. I've been very specific that nothing's changed in terms of how we think about free cash flow margins and how we look to generate the returns. We've had, as it relates to the last couple of years in kind of getting to these free cash flow margins, there's when you think about operating expenses and some of the changes, some of the dynamics in there, things that we've talked about on prior calls. Headcount has been flat to down. We've had year-over-year compensation dollars have been up as a function of a couple of things. Obviously, you're going to do merit-based pay increases. We've had a little bit of mix shift from just the percentage of employees or in certain departments to a little bit more, which are weighted towards R&D. That's naturally going to have a little bit of dollar lift on some of that. Marketing continues, as you know, we talk about our operating expenses as a percentage of revenue. That continues to click down. If you look on a comparable quarter like Q1 of this year versus Q1 of last year, or Q2 of last year versus Q2 2026, those percentages continue to click down. Sales and marketing has continued to click down as a percentage of revenue, even though marketing dollars year-over-year have been up some, et cetera. The other thing, too, to call out is we've historically not spent a ton of time talking about taxes. I know we shouldn't spend a ton of time now. We've also now effectively gotten to what is, I would call, a kind of normalized tax rate, just based on assumption and kind of geographic spend. That's been one of the things over the last year or two that's been, even though free cash flow margins have been the same, that's been one of the headwinds for us. That should largely be behind us. We feel really good about head count and the ability to continue to be disciplined there, and continue these growth rates. I'm glad we're not spending more time on taxes. April 15th is. Yeah. You and me both. Well in the rearview mirror. Yes. As is the Intuit Fireside. We're done with taxes for the day. Great. Maybe I'll close out on this. If you zoom out and think big picture, the next three to five years, what do you think will matter most in Shopify sustaining that 20%+ growth? Let's not even index to a specific percentage, but staying in the high growth category, right, which you guys have been remarkably durable at. Is it merchant acquisition? Is it better monetization? I'm going to quit giving you any more. I'll start with those two, but what would you say are the three or four biggest contributors to sustaining high growth for Shopify? Yeah. You asked me a question about monetization. Monetization will, I'm sure, to a certain extent, be a part of it, but it's really merchant acquisition, merchant retention, as a function of great products. It's a function of great products in a bunch of different areas. We spent a decent amount of time talking about international. I think that is a significant tailwind for us, and that's an opportunity to just continue to do more of what we've been doing in more geographies. We spent some time on enterprise. That is a significant opportunity. We didn't talk a whole lot about B2B. I think that is a very interesting opportunity for us because it's building a presence for us with merchants in industries historically we've not served, right? A lot of segments of B2B are used to using software that's clunky, a little painful, and we're giving them something that's very different, and we're seeing as people, as consumers engage with Shopify in various forms, and then you go back into the workforce and you think about, all right, so how do I take the brand awareness I have with Shopify and maybe think about that in a B2B context. We're seeing some interesting things in B2B. I could go on and on on cross-border, on point-of-sale, and agentic and how this plays out, which is also a potentially huge tailwind for us as well. It's a compilation of all these S curves, and how do they all, in the aggregate, continue to do more for us. That's the biggest thing. Great products will help us get more merchants and have merchants be even more successful. As we talked about Sidekick and that it truly does the flywheel on all this stuff. Well, Jeff, you guys have done a remarkable job at Shopify. I know it's great to spend some time with you today digging into it. Thank you so much- Thanks, man for your time. Thank you. We're done.