All right, I think we're gonna get started. Let me just make sure everything's working technology-wise. Yes, okay, the mic's working. So it's my pleasure for our first fireside chat for me for the day to have the team from Pinterest here as part of the conference. We've got Bill Ready, CEO. I'm gonna read a safe harbor, and then Bill and I are gonna have a conversation. So some of the statements that Pinterest will make today may be considered forward-looking. These statements involve a number of risks and uncertainties that could cause actual results to differ materially. Any forward-looking statements that Pinterest makes are based on assumptions as of today, and Pinterest undertakes no obligation to update them. Please refer to Pinterest's most recent Form 10-Q and 10-K for a discussion of the risk factors that may impact actual results. Pinterest may also discuss non-GAAP financial measures today. Refer to Pinterest's earnings press release dated July 30th, 2024, which you can find on their investor relations website, for important information about Pinterest's non-GAAP measures, including a reconciliation of non-GAAP measures to the most directly comparable GAAP measures as required. Okay, so the safe harbor's behind us. All right. July 30th does seem like a long time ago, but that was earnings, I guess, for- That's right. You guys. It has been a very long summer, at least in my world. Bill, thank you for doing this. Thanks for having me. I appreciate the opportunity to have a conversation. You know, you had an investor event last year. We've been going through 2024. To level set for the audience, why don't you walk us through your key strategic priorities- Yeah. -for Pinterest in the coming years? Yeah. Excellent. Yeah, it's been almost, coming up on right on one year since we did our first Investor Day. And maybe just to lead off, I'd say, you know, the initiatives we laid out there, the long-term plans that we laid out there, we see each of those key initiatives tracking as well or better than expected, from what our expectations were at that time, and how that stacked up to those three-to-five-year plans that we laid out, so we feel really great about that, and if you sort of dive into that, you know, I'll start with, you know, users and engagement and things that we're doing there, where you've seen now, you know, eight straight quarters of user growth from us. We're putting up, you know, record levels of new users and deepening engagement per user. So our efforts to go drive engagement with users, which has really been centered on making the platform shoppable, driving actionability. You know, Pinterest had been a place for window shopping previously. We've now made it a place where not only do people window shop, we've opened the stores. People are clicking, converting, buying, and that's not just good for the revenue side, which I'll come back and talk about, it's great for the engagement side. We see that, you know, one of the key drivers of engagement has been around, the actionability that we've brought. But then, as we drive more actionability and drive more curation behaviors, that's the second thing, so the actionability, then there's the curation that is completely unique to our platform, where we've really leaned into helping users put together outfits, put together rooms, getting more granular, where they can grab each individual item. That just gives us really unique signal that doesn't exist anywhere else in the ecosystem. In my past life, I'd have loved to have had that signal. It's one of the things that, from the outside in, I found really compelling about Pinterest. As we're leaning into that curation, that's driving engagement, but then that extra actionability, that shopping behavior and that curation behavior that's completely unique to Pinterest, that's giving us really unique signal to feed our AI, and then we're seeing huge improvements in the relevancy of our recommendations. And so on the engagement side, those have been the key points around make it shoppable and actionable, use curation to get unique signal, feed that to the AI for relevant recommendations. That flywheel is spinning, and the more we drive that actionability, the stronger our signal gets, the better our recommendations get, and there's a great virtuous cycle happening there with our users, and it's, you know, very much at the core of why we're seeing that deepening engagement per user, that's you know been a central part of the story. On the monetization side, you know, we talked about how we wanted to demonstrate that because you know, most of our users are on the platform to shop, when the user's in a commercial context, ads can be great content, right? So if the user's shopping, you know, as long as you show them that right pair of shoes or that right outfit, ad can be great content. And so that's been a central focus for us. We've really proven that out, driving up ad load while simultaneously increasing engagement per user. Really hard thing to do, but that's proving that we're making the ads great content. So we see that, you know, tracking well, and then, you know, expanding that monetization into new markets with resellers, third party bringing on more demand, all those things we see tracking really well. And you already mentioned all that comes together with our lower funnel efforts, where we've been citing that, you know, we have real strength in our lower funnel offerings, you know, which again tie back to that shoppability for the user. All those things that we laid out there, you know, continue to track as well as expected or better, and those are the key themes. Again, all those things, they come together really well because they complement each other exactly as we laid out then, and we see that continuing to not just take hold, but to. That flywheel is really spinning. Okay. All right. Really interesting stuff, and there's a lot that we're gonna follow up on that and maybe mine from that as a jumping-off point. But first, I think away from AI, the number two theme at the debate so far has been the broader macro environment. Mm-hmm. So what are you hearing from advertisers in terms of recent performance and the outlook for the second half of 2024? And maybe how do you view the competitive landscape in terms of advertisers looking across places where they can allocate budget? Yeah. So, you know, we view the macro consistently with our comments on our last earnings call. And, you know, we said there that we see, you know, a relatively stable macro. We see some puts and takes in that, but a relatively stable macro, so consistent with our prior comments on the last call. If you decompose that a bit, you know, we talked about several calls in a row now that, you know, our strength in retail, our strength in retail is really driven by, you know, the significant, you know, improvement in our lower funnel offerings and the performance that we're driving there. So we've seen that, you know, our lower funnel conversion objectives, those things continue to perform exceptionally well for our advertisers. It's a source of strength for us. So, you know, that's... Partly macro in terms of, you know, where retailers want to spend. They want to spend in that lower funnel. Great news is, we've got a great product there that's performing well, and so that's partly about us and how we're moving into that space. We've talked about how for some of those largest, more sophisticated advertisers, we're getting to, you know, five, 10% of spend getting into performance budgets instead of the, you know, sort of experimental budgets. So all that sort of addresses how we're competing to the second part of your question. And, you know, what we also talked about is how, you know, as we have, you know, really over the last two, you know, two years or so, we have really transformed Pinterest from what was previously, you know, mostly an upper funnel platform to a true full funnel with actually really great strength in the lower funnel. More than 2/3 of our revenue now comes from lower funnel. And, you know, we've talked about how we started with the largest, most sophisticated advertisers, and we've been working our way down, and we're seeing that progress as well, that, you know, those largest, most sophisticated, think of those as, you know, $30 billion-$ 100 billion in revenue or more, performing really well there. And then we've talked about how, you know, we're driving down that adoption curve as we get to the next group. And we mentioned on the last call, we're starting to get into that next group that, you know, these aren't small businesses. These are sort of $1 billion-$30 billion in revenue. We're seeing those groups, you know, really start to take advantage of our. Those retailers start to take advantage of our lower funnel offerings. And so again, that, you know, has us feeling good about, you know, performance on retail and what we can drive there with the focus on lower funnel. And then, you know, on the flip side of that, you know, we called out, you know, where there's some headwinds around food and beverage, for example, which, you know, we have a little bit more exposure to that than some other platforms, just given the use cases on Pinterest. But I'd also, you know, call out very clearly, and we did this on a call, but just to make sure it's clear, you know, that wasn't a commentary on broader CPG. It's really about food and beverage, because even within CPG, there's other places where we see bright spots, you know, around, you know, beauty and household goods and things like that, that continue to be, you know, areas of strength. Those are sort of some of the broad contours, but again, very consistent with our comments on the last call, that we see the macro as, you know, relatively stable, with some of those puts and takes and, you know, again, the strength in retail, the move to lower funnel. That's been at the core of our strategy, and we see that performing really well. Okay, very clear. You talked earlier in your answer on strategic priorities about engagement. Maybe talk a little bit deeper about where you've been taking engagement on the platform. At what point does increasing ad load have a negative effect on users? You know, you talked about ads as content- Mm-hmm. In your first answer, and this is a question we get a lot from investors. So talk a little bit about your efforts around engagement and then also tying content back to engagement. Yeah. - And content. Yeah. Yeah. So this has been, you know, I think one of the most exciting things about the business over the last two years, is that not only have we demonstrated that, you know, users will shop and buy and click and convert. It's actually not just good for advertisers. It is good for engagement. So we've talked about, you know, even as we're putting on record numbers of new users, hitting all-time highs on users, our engagement per user continues to deepen. And that's demonstrating that we're making the ads great content. We think we have a lot of runway to go, in terms of, you know, where ad load can go. It's really driven by the commerciality of the user's intent and the relevance of our recommendations. But the real proof in the pudding as to how well we're doing with that is the fact that we're able to drive deeper engagement per user, even as we take ad load up. And where you can see that, and we called it out on the last call, is our weekly active user ratio to monthly active user ratio, and that continues to improve, where we're seeing more weekly actives relative to monthly actives. Which means, you know, as we drive more of that shoppability and actionability and that curation behavior, the relevant recommendations, all of that is making the platform more compelling for our users. And so, you know, I said when I first came in that Pinterest wasn't just about, and particularly in our mature markets, about how many new users we could see. It was about how do you take those users, where we had seen most of the shoppers, you know, say, in the U.S., but how do you move them from episodic usage to regular usage? In that WAU to MAU ratio, the weekly active to monthly active ratio, as more of those monthly actives are becoming weekly actives, you know, that really says that we're making great progress there. At the core of that deepening engagement is the actionability and the relevancy of the recommendations, and that ties through to the ads as great content. You know, we've seen that be quite complementary, and we think there's a lot of runway to go. Again, it's really driven by the fact that more than half our users are on the platform to shop. They have commercial intent, and, you know, the ceiling is really driven by, you know, the relevancy of our recommendations, and we're seeing that more and more, our users see Pinterest as a place to shop and to take action, including with Gen Z, where, you know, Pinterest is where Gen Z goes to shop now, right? Yeah. And then so we see that working really well. Following up on that last talking point and bringing it back to your first answer, I want to talk a little bit about lower funnel drivers in the business. Can you talk about the success you've had with some of the lower funnel advertising tools? So I'm talking there about mobile deep linking- Mm-hmm. Direct links. You've recently announced Performance +, and how should investors think about the timeline for when positive results on clicks and conversions start to translate in increased budget allocation from those efforts? Yeah, yeah, great question. So the first thing I'd say is like, you know, we're really pleased with how those efforts have gone. Again, you know, two years ago, Pinterest was primarily an upper funnel platform, and we've, you know, really from almost a standing start a couple of years ago, stood up, you know, great lower funnel for our advertisers. I think, you know, we share the urgency of how fast we want to move on those things, but you, when you sort of step all the way back. You know, mobile deep links, which was our first, you know, our first product that really was driving click and conversion for advertisers, that only went GA just over a year ago. That was, like, middle of 2023 that went GA, and that was really just for the very largest advertisers because it was a deep link into native mobile apps, well, it's really just the largest retailers that have big penetration on their native mobile apps. We saw that working really well, sort of middle of last year into Q3, then we said, "Okay, how do we take it to the next group of advertisers?" And that was Direct Links, where we said, "Oh, well, now this can work for even those that don't have a native mobile app." In parallel to that, we launched Conversion API, where, you know, if you're an upper funnel business, you don't need Conversion API because you're not driving conversion, so and you don't need lower funnel measurement. So we launched measurement, and we started driving real adoption around measurement. And Direct Links, to keep in mind, like, Direct Links didn't get to full rollout until Q1 of this year. So, you know, we're still quite early on, even though we've seen tremendous momentum and, you know, we're driving through that adoption curve, and each one of those things sort of adds to our penetration there, both the penetration of the relevant inventory we can bring onto the platform so that the user is seeing the right shoppable inventory from all the retailers they wanna buy from. And so, of course, we started with the largest ones that could bring the most inventory, but we're expanding that now from those 30 billion+ retailers to the billion to 30 billion retailers. Seeing traction there, and that really is driven by, you know, the compounding effect of, you know, being able to drive clicks and conversions. We've talked about three consecutive quarters in a row, more than doubling the number of clicks we sent to advertisers year- on- year, really driven by mobile deep links and direct linking. You know, we've talked about driving adoption of our measurement tools, our privacy safe measurement tools, that lets the advertiser measure it. And then, on this most recent call, we talked about the launch of Performance+, which is, you know, our AI-driven suite that makes it so that that next group of advertisers down that, you know, has sophistication, but, you know, they, they, they need these things to be simple. It makes it so that, you know, we've driven the performance, the clicks, the conversions, the measurability, and now we're driving easy campaign creation and setup, right? So with Performance +, things like, you know, automated bidding, budget allocation, targeting, the dynamic creative optimization, where we will help optimize their creative. You know, it's still in beta. These things have a multi-year, you know, adoption cycle, but we're seeing really great results from that. And, And, you know, on the last call, I called out, you know, advertisers like, Poshmark and Timberland, which would be in that sort of next group down, right, as we're driving further down, in the adoption cycle, and how they were seeing great results in our beta. You know, we've also had that broadening continue with retailers like, you know, Tractor Supply Company that from our dynamic creative optimization has seen their return on ad spend double for the ads using our dynamic creative optimization, where we can take our unique signal of what's relevant to the user and help optimize the creative to get better performance for the advertiser. But also, it makes a more visually compelling ad for the user, so again, the ads are great content. You know, so we see all that coming together, but if you just step all the way back from it, you know, we are well along our way in terms of proving out the theory of the case, right? That we have broken into performance budgets. We're getting, you know, 5%-10% of total spending, or 1% of them across the market. But for, you know, some of those largest, most sophisticated, we're getting 5%-10% of budget from them, and as performance budgets instead of experimental budgets. Now, we're taking that to the next group of retailers down, and all these things, the, you know, mobile deep linking, Direct Links, the measurement, and now Performance+, these all have a compounding effect, where they drive us deeper and deeper into that adoption. And, you know, while we've got very clear proof points, really great results, we're still early innings in terms of the yield that we can get on that. And even for the very largest, we're still getting more yield. We're opening up new categories and those kinds of things, where they're bringing more inventory on. So we're still seeing great yield from the largest ones, even as we're now getting to those next tiers of retailers. And so, you know, we see a lot of adoption cycle in front of us, even as we have really great proof points that say, "We're a real player in performance now." Okay. You've announced a number of third-party partnerships over the last 12, 18 months. Maybe pivoting to that as a topic, what are your key learnings across those partnerships? What inning are you in, in terms of scaling and revenue contribution, and how do you see those partnerships evolving over time? Yeah. So, you know, we announced our first third-party partnership with Amazon a little over a year ago, sort of early 2023. So we're a little over a year into that, and it's performing exactly as we expected. We said when we announced it that we wanted third-party demand to round out gaps in our auction and bring more shoppable, relevant inventory for our users, and we see it doing exactly that. And we've seen nice, steady progression of that. We announced our first international partnership with Google earlier this year. And so we see, you know, nice, steady build of those things. It's still relatively early innings with a lot of opportunity left to go, but we are continuing to see that stairstep up as we talked about on our last call. And, you know, we're expanding further, expanding into other categories in our mature markets and seeing continuing yield from that. But we're also launching new markets. So, we are now starting our international expansion of the Amazon 3P partnership, starting with Canada and Mexico. And so, we see good opportunity on the international front, but again, we're still getting a lot of yield out of our largest, most mature markets. But, you know, across each of those dimensions, we see our third-party partnerships overall doing exactly what they were intended, which is rounding out gaps in our auction, and helping us do more in unmonetized or under-monetized markets, as we talked about before. Okay, understood. I wanna turn to artificial intelligence. You guys have been at the forefront of a lot of AI and machine learning. In my opinion, you know, things like visual search and computer vision jump out. Can you walk us through how you're leveraging AI from a user and a monetization perspective, and touch more on the broader long-term strategy around AI for Pinterest as a platform? Sure, yeah. You know, AI is a core competency for us, and we've seen really tremendous yield from it. Just to put this in perspective, just from things we talked about before, you know, we talked about when we moved to GPU serving large models, you know, that was a little over a year ago that we started that. You know, we saw that we got, you know, 10 percentage points lift in the relevance of our recommendations from that. Part of that is about the power of large models and GPU serving. Another part of that, though, is about the really unique signal that we have. And so that's part of our AI strategy as well, which is, in addition to, you know, using large language models, tuning them to our unique signal, we're leaning into the things that gain more of that unique signal for us, like the curation behaviors and things like that, where when people come onto Pinterest, we don't just see what they clicked and bought. We see them days, weeks, months before they buy, where, you know, they're thinking about, you know, how to update their wardrobe for fall. We see them well ahead of fall shopping, starting to put outfits together and say, "Okay, well, yeah, I'd put, you know, this dress and this handbag and these shoes together. I'd put, you know, you know, this jacket with these pants," and we see hundreds of millions of users that are making all these product associations that computer vision alone wouldn't get you because it's not just pattern matching, right? You don't want your dress and your handbag and your shoes to be the same pattern necessarily. That doesn't make a great outfit, but what does, you know, make a great outfit is being able to see, oh, here, here's how millions and millions of users are thinking about pairing this kind of dress. What kind of handbag and shoes would they put with this dress? And then how does that line up with this user and the recommendations we make? Those things are about us really, you know, capturing more and more totally unique signal. And so I think as we project forward with AI, you know, there's... I think you've got to sort of break AI into sort of two components. There's one part of this that is, you know, analogous to what happened with the cloud compute providers, where it's sort of a clash of the titans that, you know, in terms of how many, you know, cloud providers were there gonna be, you know, you're gonna have a few that would, you know, be the major providers of cloud infrastructure. But then you also have lots of people who would benefit from consuming that cloud infrastructure and those building blocks. And we think about AI similarly, where if you think about the building general-purpose large language models, that's massively expensive. It's sort of a clash of the titans and, you know, sort of trillion-dollar market cap companies or those that are very closely aligned to trillion-dollar market cap companies because it's a lot of spend, but then what we're seeing is that we can go take those large language models and tune them to our unique signal and see significant increase in yield from that signal, so to make that really tangible, you know, when we test this across, you know, most of the major model providers, we're always testing those. And we see that when we take those large language models, the general-purpose large language models, bring them in-house, and then tune them to our unique signal, you know, we see, you know, 300 basis points, you know, lift from tuning to our unique signals. And to put that in context, you know, this is a world where, you know, in past life, you know, if you got 10 basis points improvement, 15 basis points improvement, like, oh, my gosh, you're doing cartwheels, it's fantastic. How powerful is our unique signal? 300 basis points improvement over what we would get from those large language models off the shelf from those providers when we tune them to our unique signals. And then that unique signal is getting higher and higher fidelity as we get more engagement per user, more shoppability, more curation and action on the platform. We're just getting totally unique signal that others wouldn't have, and that's, you know, evidenced by that 300 basis points lift, that we get when we tune to our signals. And so, you know, in that, you know, I think you've seen us really, really demonstrate AI as a core competency, where we have some of our own deep computer vision stuff. It's, like, really narrowly fit for purpose, where we do some of our own models just for these things that are really fit for purpose. But then we're taking the off-the-shelf large language models, but getting totally unique results from them because we have totally unique signal to feed to it. Stated very simply, the AI is only as good as the signal upon which it's acting, and that's why we're really curating and harnessing this totally unique signal that we get from the human curation on our platform, which I've talked about a lot. But that's just another example of quantifying that: a 300 basis point difference in the relevancy of the recommendations based on how that model would have performed off the shelf versus how it performs once it's tuned for our unique signal. So we're continuing to lean into it that way. Okay, so that's an example, really interesting, about how you can take AI and drive a lot of improvements in your own business. A pivot on that, that's probably one of the biggest debates I get or questions I get from investors is, in a world of digital advertising, you are competing with other companies that are either hyperscalers or- Mm-hmm. Spending money like they're hyperscalers. They're spending tens and tens of billions of dollars on AI. Maybe bring it back to the competitive landscape. Yeah. So how do you see AI, the propensity to spend, versus bringing it back to competition and where Pinterest fits into that worldview? Yeah, I think on that, you know, a couple of things I'd say. One, I think we've proven out that we can compete really well. Like, we are taking share in performance budgets. And I think that really gets to just how much the uniqueness of your signal matters, right? And the inherent intent of users on our platform, right? You can't sort of. It's very hard to change what the user wants to do in your app. And, you know, this is one of the things I saw from the outside looking in before I came to Pinterest a little over two years ago, was that the users were here to shop, but they just didn't have the ability to take action. You know, they were window shopping, but all the stores were closed. As we've opened those stores, they're taking action, but then this curation signal is a totally unique signal. And so, you know, when you ask, like, "Bill, how are you breaking into these performance budgets when, you know, you were sort of at a standing start two years ago?" It's because we have this totally unique signal around the curation behavior that lets us make these really great recommendations. And so if you go talk to users on Pinterest, and ask them, "Well, why are you going to Pinterest?" Right? You know, one of the things they will say is that, "Well, Pinterest just gets me." And, you know, I talked about that, you know, 10 percentage point improvement in the relevancy of our recommendations when we first went to large language models, you know, roughly a little over a year ago. That is very user perceptible. After we did that, that's when we started getting feedback from users that were like, "Pinterest just gets me. Like, wow! Like, it's giving me great recommendations, it knows my style, it's helping me discover new things." And that's very different than what has existed elsewhere, where, you know, there's lots of other places that, you know, when the user knows what they want, you know, they'll go, you know, search for, you know, Nike Air Max 270, you know, size 11 in stock now. Well, okay, those are things like the user sort of decided what they want. And I think the last 25+ years of e-commerce were really about when the user knows what they want, help them get it the cheapest and the fastest. I think this next decade is really gonna be about the rest of the shopping journey, which is people start with this much more general sense of, "Help me find some cool running shoes," and from that, because we have our Taste Graph and that unique curation behavior where we don't just know that user's taste, we know how other users styled those sneakers, you know, what did cool running shoes mean to them? What were the different brands available? That's where we've just got really unique signal that's letting us give great shoppability to the user in a very different way than they experience anyplace else, because the user has intent, but it's this loosely defined intent, which is a magic moment for the advertiser to meet the user when they have clear intent but haven't yet decided what to buy. That's a unique moment in the user journey that we have, but then unique signal we get from that, that lets us drive really great performance. So it's not a bet, it's not a bet about can we beat, you know, trillion-dollar-plus companies on the raw horsepower of the AI. My bet in coming to Pinterest was that that raw horsepower, just like cloud compute, was going to be broadly available. Just like cloud compute, those building blocks would be available, and then who has unique signal to go do something? Who can assemble those building blocks in a unique way? And we have a user behavior, user intent, and really unique signal that let us just go do something different with that AI, and we're proving that out quarter after quarter after quarter as we gain share in those performance budgets, capture share there. I think it's also. It's not even that. I think when you look at our user behaviors, so for, you know, talk about the advertiser lens. When you look at it through the user lens, I don't think it's even that we're competing for existing behaviors. I think it's more blue ocean, where, you know, the first 25 years of e-commerce solved for buying, but not really for shopping. With shopping being the, "Hey, I-" You know, think about how you shop in the physical world. You want to refresh your wardrobe for fall. Whether you walk into a mall or a series of boutiques, you have sort of a general idea, and then you sort of walk the bazaar and discover and peruse. That's what Pinterest does in this purely visual search kind of way, and it's unique behavior, and it's, I think, letting us compete really effectively. Okay. I know we only have a few minutes left, so I'm going to try to squeeze one more question in. You talked earlier about Gen Z as a percentage of your base, I think 40% now of your base, fastest-growing demographic. Why do you think the platform is resonating with Gen Z? Yeah. So this has been, you know, one of the brightest of bright spots for us. I think two years ago, when I was coming in, I was getting questions like: "Oh, is, is Pinterest even going to exist?" you know, sort of the narrative was it was aging up and aging out. And, you know, our app is actually aging down now, which is really unique, you know, for, for mobile apps. that, that over time, you know, you tend to sort of age up, and we're aging down. You know, Gen Z is now more than 40% of the users on our platform. They're the largest, fastest-growing demographic on the platform. We are growing across all demographics that we track and all geographies that we track, but Gen Z is our largest, fastest-growing. At the core of why we're winning with Gen Z, you know, there's the shoppability and actionability that I've talked about, that is part of it. But Gen Z loves the curation behaviors, right? They really, you know, Gen Z, they're at a time in their life where they're still figuring themselves out, right? And so investing in that time to go, you know, what's their style? What do they want that to be? You know, they're getting their first apartment. Well, how do they wanna design that first apartment? What do they want? They spend a lot of time curating, and they love that. And then in addition to that, you know, we've talked a lot about investing in Pinterest as a more positive alternative to social media. And if you ask Gen Z why they go to Pinterest, you know, two things they will pretty reliably say. One, they'll say that they see it as an oasis away from the toxicity they experience elsewhere on social media. Stated simply, you know, social media is where they, you know, where there's lots of engagement over their selfie. Pinterest is where they go to invest in themselves. You know, social media is about their selfie, Pinterest is about investing in and figuring out themselves for themselves, and sometimes with a small group of, you know, close collaborators, close friends, or people of like interests, but it's a very different place than the rest of social media. And so they really love that positivity, but then the curation behavior and the shoppability is, you know, quite central to that as well. I think that's worth noting that, you know, we've got these things that are working for Gen Z, but they have broad applicability. You know, because the curation is broadly applicable beyond Gen Z, the shoppability is broadly applicable beyond Gen Z, the positivity, broadly applicable. It's not just good for the user, it's great for advertisers. Brands wanna be in a positive environment. And so we see that really cutting through, and again, it's why we're, you know, Gen Z's our largest, fastest-growing demographic, but we're growing across every demographic, every geography that we track. And I gave that stat on, you know, on weekly active to monthly active, and that deep engagement, we shared at our Investor Day a year ago. It's also exceptionally rare for an app to age down. Our most recent user cohorts that we're bringing on are approximately twice as engaged as cohorts from years prior. So we truly have our best product market fit ever on the platform, and it is all these things together working really well. And so that, you know, weekly active to monthly active, where you're seeing more of the monthlies become weekly actives, that's broadly applicable, too. And really important to call out, it's the most true in our most mature markets. So we see the best weekly active to monthly active ratio in the U.S., and U.K. and Canada, and we see that across our mature markets as well. So all these things line up, you know, quite well, and I'd say just, you know, as we come up on time here, one of the things I didn't mention in the AI comments, I think part of us having AI as a core competency, you've seen us really simultaneous to transforming the product for shoppability, you know, standing up a performance ad platform from scratch. We've also driven great margin expansion. We've put great discipline in the business and all those things where we're lining up, you know, really efficient use of AI, aligned with things that drive clear user engagement, clear monetization. We're getting good unit economics on what we do with AI. As we're building the business, you know, we're opening up new demographics, driving deeper engagement, getting great use of AI, but also making Pinterest just a much more sustainable, durable business, great margin expansion, and finding ways to get great yield out of our investments in the product. Okay. We are gonna have to leave it there. Bill, thanks so much for being part of the conference. Thanks for having me. Please join me in thanking Pinterest for being part of the conference.