Match Group, Inc. ($MTCH)

All right. I think we're getting started. Good afternoon, everyone. Cory Carpenter, Internet analyst at JPMorgan. Pleased to have Steve Bailey, CFO of Match Group. Thanks for joining.

Happy to be here.

So starting with the safe -- Match's safe harbor. During this presentation and during the question-and-answer session, we may discuss our outlook and performance. These forward-looking statements may be preceded by words such as we expect, we believe, we anticipate or similar statements. These statements are subject to risks and uncertainties, and our actual results could differ materially from the views expressed today. Some of these risks have been set forth in our periodic reports filed with the SEC. Also today, we may discuss certain non-GAAP financial measures. Reconciliations to the most directly comparable GAAP financial measures are provided in the published materials on our IR website. These non-GAAP measures are not intended to be a substitute for our GAAP results. Are you timing me? Okay.

So let's see you go on to the questions. So let's start with, I mean, I think it's no secret the dating industry has been challenged in recent years. So what's your view on the state of the industry? What would you tell investors who may still be skeptical on the growth outlook for you?

Yes, it has been challenging as a category as a whole. It's not unique to Match or unique to Tinder. It's a category problem. And I think it's a product problem at that. Really, what's happened is the needs and tastes and wants of Gen Z, which is sort of the new generation of daters, is different than what it was with millennials. And as we look back, I don't think we evolved the products fast enough, particularly Tinder, to meet those changing needs. We kind of fell asleep at the wheel a little bit, and now we're playing catch up. The good thing is what we know from all our research is Gen Z wants meaningful connections. They want to meet new people. They just want to do it differently. They want it to be lower pressure more than anything. They want to sort of vibe their way through it. They don't want it to be a job interview. What Tinder was novel, cool, part of the zeitgeist for millennials, it's not the same for Gen Z, and we sort of level up our game, and that's what we're hard at work doing.

So even it Match through, I think, 5 CEOs...

5, 6.

6 CEOs. So a number of turnaround efforts at Tinder. What feels different this time around that gives you the confidence that you're really turning the corner?

I would say 2 things. One is leadership. I mean, it starts at the top. And yes, I've been through -- Spencer's my sixth CEO. They all have their strengths and weaknesses. But I think Spencer more than anything is by far the most experienced. He's the first, I think -- yes, the first real public company experienced CEO. He ran Zillow for 10 years. He's also a founder, right? He founded Zillow, he founded Hotwire. He's got a lot of sort of chops, street cred and a lot of experiences to build up over time. And how that shines through is a level of decisiveness. And he brings certain principles that he's came in day 1 with that he's now executing, like a focus on user outcomes, for example. His whole -- the whole notion of it starts with building the right team that ends with the stock price, not the other way around. And so because he came in with a effectively, a mental framework for how to run a tech company, he's just been able to move very fast with a lot of confidence. So that's different -- clearly different than before. And then the second thing is the strategy. And so our focus away from purely revenue and profit and EBITDA, which we're very good at. Tinder's very good at monetization more towards user outcomes and the user experience. I think gives me confidence that we're building this turnaround from the ground up. And so that means it has staying power.

Okay. So we'll spend a good chunk of time here on Tinder. I think, first, high level, you framed that Tinder turnaround is a 3-step process, reset, revitalize, resurgence. Could you elaborate on these 3 phases and where you're at in the process?

Yes. I think reset is people and culture, and that was the focus of 2025. We did a little bit more of it in Q1. We merged Match Group Asia with E&E, but that's largely behind us. And so to my early point, in Spencer's framework, it starts with the right people, the right team, and I think we're largely there. The revitalization phase is what we're clearly in the middle of now, which is all around product innovation. So now that we've got the right people, the right team, we've restructured the management layers. We've built a culture of urgency and accountability. Now we're building products, particularly at Tinder and at Hinge. And so what we think is through product innovation, that will get us to the resurgence phase, which is back to growth at Tinder and for the company as a whole in 2027. So what we've said is revenue growth and payer growth by Q4 of 2027. That's the goal we set. And I think we're on a pretty good path to get there.

Okay. So a few quarters ago, you started giving stats showing Spark's metrics improving. I'll let you explain what that is. But you built on that this quarter with registrations and retention both returning to growth for the first time in a couple of years. So what's driving these improvements? And how do you think about the sustainability?

Yes. So Sparks are 6-way conversation. That's our proxy for a meaningful connection. That's the focus of Tinder and the company. No longer is it a focus on just getting people likes that we can monetize. It's actually getting into -- getting users into real meaningful connections, which is why they're on the app in the first place. So that's the focus and everybody asked for. And we needed 1 internal too, like a metric that we can track progress against, and that's where we came up with Sparks and Sparks Coverage. And for the last few quarters, that's been improving, which is great to see. You're right. Sparks Coverage, which is of the users we have, how many of them are getting into meaningful connections, that was up about 6% in Q1 year-over-year. And even Sparks itself, which is the volume of 6-way conversations we have, which is basically a function of MAU times the coverage, is nearing flat year-over-year. It was down 1% in Q1. So we're making a lot of progress on the early indicators. And then the question was really what will that sort of carry through to the more core financial metrics that investors really care about. And I think the answer to that is yes. The metrics, you mentioned a couple, retention, which is a really hard metric to move for our business and most sort of consumer tech businesses is up 1% year-over-year globally. It's up 3% year-over-year for young females in the U.S., which is our core demographic we're going after. And so what that's saying is the thesis is proving out. That more Sparks and better outcomes for our users is leading to better user retention, lower what we call bad churn, less people leaving because they're not finding success on the app. And actually, it's a -- and what that's leading to, combined with improving registration trends, is while not yet positive, improving MAU trends too and payer trends as well. So the thesis is holding out, is bearing fruit. And I think we're probably a little bit ahead of where we thought we'd be, but we're clearly not at the finish line yet.

Okay. So you -- Tinder hosted their first-ever product event a few months ago. I was there. Very nice.

Thanks for coming.

Where -- and you announced a number of new features there. Which of the product launches have been the most impactful thus far? And what can you tell us about the product road map in the second half of the year?

Yes. Actually, the single most impactful change we made was with algorithms. So more on the back end. And what that meant was largely moving away from algorithms that optimize for likes distribution to moving towards algorithms that operate -- optimize for outcomes for Sparks. And we've seen a lot of success with that. I would say that, that's 2/3 of the benefit we've seen is really in the area of algorithms, which I think does bode well for the sustainability of the turnaround in that that's so core and fundamental to the product. It's not a fancy bell or whistle that gains of excitement that fizzles out. It's the core crux of the product experience. When you combine that with a new marketing strategy, which is a focus on further down the funnel, not just brand marketing, and new innovative features like Double Date, for example, that's really resonating with Gen Z. 1 in 4 Gen Z females in the U.S. is using Double Date, which is -- for those of you who don't know, pairing up with a friend, swiping all the pairs and going out in the double date. It'd be good. The combination of those things is leading to sustainability of the turnaround.

Okay. So I told you this, my goal of today is for everyone to walk out understanding the giveback concept. So we can discuss it now. Okay. We've got a lot of questions on this. So you've talked for the concept of a $60 million user give back budget for this year? Because you didn't use as much as you thought you would use in the first half of the year. Could you just clarify exactly what this is and why you did not end up utilizing those givebacks in the first quarter?

Right. So at the beginning of the year, as part of the strategy around optimizing for user outcomes, we realized that some of the changes we were going to need to make to the product could have a short-term negative impact to revenue. So for example, the algorithm -- the example I just gave. If we were concerned that if we change the algorithms that were effectively optimized to maximize revenue through likes distribution towards one that optimize for outcomes towards Sparks, we would lose revenue in the short term. But we were prepared to do that because we knew monetizing declining user base is not a winning strategy. Turning around users, getting it back to growth is a more sustainable strategy for long-term revenue and profitability. So that's why that was the reason behind setting up the $60 million giveback budget. Now what we found is we started testing these things. Most of the tests, a lot of tests have centered around algorithms that -- the outcome of that was not as bad as we expected. I'll give you just an example. We tested 2 big algorithm changes. First of all, we have multiple algorithms and not just one. We've tested 2 big algorithm changes. One kind of played out like we expected. It drove higher -- more Sparks and better Sparks Coverage, but it cost around $15 million in annualized revenue, negative revenue impact. That's kind of what we thought would happen. But there was another algorithm change that we made that drove improved Sparks and Sparks Coverage but actually also added $30 million of annualized revenue, right, which was unexpected. So when you net all these tests together, what we were left with was rolling out tests that improved user outcomes and actually added to revenue, not detracted from revenue. Now that's why we didn't use as much of the budget as we expected in the first half of the year. So what we've done is while it's been lower than expected in the first half of the year, we're saying we expect that to increase in the second half. So we have $45 million budgeted for user givebacks in the second half of the year. And the reason for that disconnect is because what we plan to test in the back half of the year is not the same thing as what we test in the first half of the year. It won't be centered as much around algorithms. It will be centered around other feature and UX changes that we went out to test. And because we're unsure of the impacts of those tests, we're keeping that budget there. And we -- there are a few that we're ideating on that are pretty bold tests. And we don't want to constrain the product team for making really bold product tests that improve outcomes and improve the user experience because they're worried about hitting the revenue guidance. So we've left the $45 million cushion in place to see how that plays out. But if the past couple of quarters is the predictor of the future quarters, there's a chance we could come under in the back half too.

Makes sense to me. If anyone still needs clarification, feel free to raise your hand.

Thumbs up, thums down.

All right. So this is probably building on that question. But the guide you gave, it assumes the $45 million give back in the back half of the year. It also assumes that Tinder -- you don't guide formally to Tinder payers, but you're roughly saying Tinder payers kind of stay around the negative 5% in the back half of the year. So why would we -- I think the question we get is payers improve from down 8% to 5%, why would we not see continued improvement in that metric?

And that's because of the user givebacks. Because of the back-weighted nature of the user givebacks, if we end up utilizing that $45 million, which again, on the P&L shows up, it's $45 million less revenue, which flows through after IP fees down to EBITDA, right? And it also shows up in the payer number. Because that's where -- it's not really a pricing or RPP thing. It's more a payer thing. We're giving a better experience for the collective user base, most of which are free, and for the ecosystem as a whole, but it could be -- but it could cost us some revenue and payers in the short term. And so if you did the -- think about it as the impact of payers is proportional to the impact of revenue. And so if you think about the $45 million as a percentage of the second half revenue for Tinder and take that as a percentage, think about it similarly on the payer side. And so if you play that math out, basically, what's happening is the improvement in MAU and overall business trends that has been leading to improved payer trends will continue, but could be offset by the user givebacks. And that's what's netting us out to about down 5% each quarter. If the user givebacks don't happen, then we should continue to progress more towards flat by the end of the year. Still probably down, but meaningfully better than the negative 5% we are at today.

Makes sense. Okay. We'll move on. So let's talk about AI a bit. I think, first, from a product perspective, how is AI impacting the Tinder user experience? And what role do you see it playing in dating more broadly?

I think it has broad applications. The few places that it's showing up in Tinder is one, algos, which we've talked a lot about. It is so much better than the old rules-based algorithms. It can use more unstructured data, more nuanced data. The word is probably inference. It can infer things about you that the old rules-based algorithms could not. So we found a lot of success there at Tinder and at Hinge. And I think it's still early days. And at the -- like I said before, the crux of the product experience is the algorithms. So that's one. The other one is more forward-facing features. And Chemistry is a good example. Still pretty early, but that is a curated drop or match, highly curated match that's completely different than the core swiping experience. So think of it as AI gets to know you a little bit better through a like or late week back and forth. And therefore, because it knows more about you, it can give you a better match. And that's something that would -- is resonating with Gen Z too. They don't want to just get on the app and swipe. So that's another area. And then the third area will be trust and safety. There's a lot of really co-applications with AI around trust and safety like Are You Sure? Does This Bother You? These are nudges within the app that help improve the trust and safety experience across the app.

And then just from a cost perspective, a lot of conversations this week around token costs and all of that. So how do you expect AI to impact headcount and expenses across the company?

Yes, we're embarking on a huge AI enablement push. We've got our AI Day internally this week. We gave -- we announced a week ago now that every employee in the company would get access to cutting-edge AI tools. So we think this is a big opportunity. And the way we're effectively paying for it is by slowing down hiring. I think you see that across the tech industry. Folks are taking 1 or 2 paths. I mean, I think for everybody, the AI costs are coming in higher than they expected. You hear that time and time again. We did not budget for what could be $5 million to $10 million in tooling costs for our business on an annualized basis, which is meaningful. That was not in the budget. But we see the benefit of it. So the way we're paying for it is effectively slowing down hiring. Companies are either slowing down hiring or they're reducing the size of their headcount. We're doing -- we're slowing out hiring instead. We had gone through a big RIF last year. We feel pretty good about where our headcount is. And so that's effectively how we're paying for it. The reason is twofold. One is not only because we need to sort of find a way to pay for these tools, but also because we want to see the needs of the organization, what roles do we need, what roles we don't need as we infuse AI through all the departments of the company.

So sticking with expenses. You outlined a few areas of cost savings at earnings a few weeks ago, you have app store fees or a good tailwind for you this year, assuming it don't change again. Maybe could you talk about your -- the big expense initiatives that you're doing this year? And I think more broadly, like how much more -- you're pretty profitable, 40-ish percent margin. So how much more opportunity do you see on the expense side?

Yes. I think there's 2 pockets that I'm focused on right now. We generate about $100 million in savings on the headcount side. We've generated over $100 million in savings on the IPP side. So I feel good about -- I think that's pretty optimized, and we fed that back into the business for further investment in growth instead of dropping it down to the bottom line, but it gives us a lot of flexibility. The areas we've spent a little less time in, one is cloud costs. So we've got an initiative particularly at Tinder to reduce cloud costs through dev optimizations. That was a co-base that was built up over 10 years, really hadn't been a focus. So it's not going to be as big as IP fee savings. Our cloud costs are, round numbers, call it, $100 million per year, but I think we can chip away at that. We've got a plan to do it. And the teams are executing against some goals there. And then the second probably bigger one is really around marketing. And so we talked about in the last call a couple of weeks ago that we merged the performance marketing team into 1 team that sits across all brands of the company. We spent $600 million a year in marketing rough numbers. It's been done in a somewhat siloed way. So now we've got a whole new approach to marketing, which we call Prism, which is a framework we use that shows marketing ROI on an apples-to-apples basis across all our brands. And then we've also now merged the teams into 1 performance marketing team that I think will unlock a bunch of efficiencies. The way I think about it is can we optimize the $600 million by 10%, right, given the fact that it was kind of run by -- in a siloed way before? Probably. So that's $60 million now you can drop that down to the bottom line or you can keep spending the $600 million, but get 10% more for it, 10% more [ regs ], 10% more payers, 10% more revenue. We'll probably go that direction because we think marketing is working for Tinder, but I think that's a big area of optimization for us.

Okay. So moving to your other brands. So Hinge continues to track towards your $1 billion revenue target for next year. Where do you still see the most opportunity for growth at Hinge?

I think there's really 2 big -- there's 3 big pockets of growth. One is monetization in core markets. They're still undermonetized relative to Tinder if you look at it on an apples-to-apples basis. I know folks get confused, like the RPP of Hinge is so much higher than the RPP of Tinder. You're only looking at 1 side of the equation really. It's really RPP and payer penetration that matter, both of them matter. And if you just look at revenue per MAU for example, which nets the two together and control for the geography mix and the gender mix, there's still runway for Hinge to monetize better in core markets. I think bringing a better value proposition to females is another big lever for them in core markets where females are undermonetized. That's a category-wide thing that they have some real interesting plans to address. Then the second is just more room for growth in the existing expansion markets. Europe is still -- while they've been there for 2 years, it's still a pretty small percentage of total revenue. And if you just look at it in proportion to the U.S., we've got other benchmarks like Tinder, for example, there's still a long runway to go there. I think that will come naturally through just sheer momentum. I mean, the revenue was up 100% in the last few quarters there year-over-year in Europe for Hinge. And then the third is expanding to new markets. And so I think what I'm excited to see is not that South America is a massive market from a revenue perspective, but what it's showing is the continued global appeal of Hinge as a product. And what that tells me is it builds my confidence in can we take it to Asia? Can we take it to bigger countries like Japan, where there is a lot -- really high TAM and revenue base? I think the answer is clearly yes there, and that will provide a long runway too. We don't need Japan to get to $1 billion, but we needed Japan to get well beyond $1 billion. And I think we're all really confident in the progress we're making there.

So I think one question we've started to get now that Tinder is stabilizing is, do you have any concern that a Tinder resurgence could perhaps come partly at the expense of Hinge?

No. I mean, what we've seen is cannibalization has not been an issue. It's, again, a category issue. When we enter new markets with Hinge, like when we entered Mexico and Brazil, we looked at it really closely to see is Hinge an early success there, taking away from Tinder? And the answer was clearly no. Tinder's trends held up quite well. Hinge added downloads, and Match Group as a whole, grew in those regions. And I think the opposite holds true. I mean, one of the things we're really focused on is the [ gem ] we talk a lot about, which is where our brands fit within our portfolio. And clearly, hinges in the focused part of the gem and directly across from that is the fun part of the gem, which is where Tinder lies. So as long as we keep those products distinct and the brand positioning and the product positioning unique, I don't see a problem. It's a multi-app usage category. Users using 3-plus apps at a time, I think there's a use case for both. So I mean, that -- what we want to do is keep the Hinge growth going and get Tinder back to user growth. That's the formula to get the Match Group stock price up and the company back on the right footing.

I want to ask about -- you've made 2 minority investments. Sniffies is very recent, and then also Overtone from the Hinge founder. Could you talk about these 2 businesses and just how you're thinking about the -- why the minority investment route and how you're thinking about the optionality to build that ownership over time?

Yes. There are a little bit different use cases. Overtone, we made a minority investment in -- or sorry, Sniffies is we made a minority investment in, was a founder-led start-up that we've invested in. I think the reason behind that is it's the biggest segment of the category, the non-heterosexual male segment of the category that we don't really play in. And we try to do that with Archer. They build a great product. It's been a couple of years. We've invested a bunch of marketing and in people, and it hasn't gained the product market fit that we had hoped. So we pivoted the strategy. We've shut that business down. We're going to save on OpEx, on headcount, on SBC, which helps the P&L, and we're making a minority investment in Sniffies, which is the #2 player in that space. And we think a path to #1. They'll benefit a lot from our trust and safety know-how, just our institutional knowledge of the category. But at the same time, a minority investment does a couple of things. One, it keeps it off our P&L, although it's a profitable business and growing quite nicely from a revenue and EBITDA perspective. But it also lets them continue to be a startup founder-led organization. So it doesn't sort of lose the magic created by coming in and being part of a larger public company. And so the idea is leave it founder-led, give it more support, the support that it needs to reach the full potential, and then we have the option to buy it down the road, which is a similar approach to what we took at Hinge, and that's worked out phenomenally well. We bought Hinge for $20 million, $30 million back in 2017, and look at it now. So if we could repeat that playbook, that would be great. Overtone is a little bit different. Overtone, we incubated internally and then spun out, and we have a minority investment there. That is because it's a new bet on the category, right? It's a truly AI-native experience. We think there's room for that in the category. Nobody has really cracked the code yet. I think Justin, given his experience in the category, has a leg up against other start-ups. But it also is going to take a while, and it's going to accumulate losses and there's a low probability of ultimate success. So I'd like to do those things sort of off balance sheet in a minority investment way with the option to buy it. To the extent it does gain product market fit, we think it's a good long-term fit for the portfolio.

I think that's a good segue to just capital allocation. You kind of laid out your capital allocation framework at your Investor Day 2 years ago, maybe at this point. Maybe refresh us on...

December 2024.

Yes, refresh us on the capital allocation strategy? And are there any changes that you're making to it?

The short answer is no. Look, I think Sniffies is a big bet, but $100 million out of the $1.1 billion in free cash flow for the year, like from my perspective, it's very manageable, right? So doing a Sniffies deal does not take away from our commitment to return 100% of free cash flow to shareholders through largely buybacks and also a dividend. So that hasn't changed. And I think going forward, we don't have any plans to do large-scale acquisitions. We'll continue to do this tuck-in approach. But the primary objective will be -- really, our first and foremost objective is invest organically in our businesses, right? So we're investing in marketing and a product that's [ inter ] and Hinge to keep Hinge growth going and a turnaround at Tinder. We feel like they have the right levels of investment. Second is returning cash flow to shareholders through buybacks and the dividend. And I think it's a really powerful formula given where the stock price is at, we bought back about 5% of shares over the last year. So we're buying down the shares outstanding, while turning around free cash flow. Those two things compound on each other. And free cash flow per share, for example, grew, I think, 23% last year. So that's a powerful formula if we can keep it going.

Before I ask about app store fees, any questions in the audience? All right. [ Bore you a store ]. Okay. So you've talked about, I think, over $100 million of savings from lower app store fees this year. I think two questions. One, now that you're kind of fairly advanced in your rollout of alternative payments that would have been your key learnings? And then secondly, what are the key pieces of legislations or trials that you're watching that could impact fees in the future?

Yes. It all comes down -- a couple of learnings. One, friction is the most important thing. So as we've improved the experience to make it more and more frictionless, that has led to better financial results. And as we said, we're about 10% ahead of expectations against that $110 million in savings. So a lot of that is through is just optimizing the flows. Like 1 win was just improving the tech around web payments that increase the time, the checkout time, fractions of a second, adds up to real dollars. And then two, I mean, using Apple Pay, payment options that are pretty frictionless because it's different than -- Apple Pay is different than IAP. Apple Pay is more like a credit card. But the adoption of Apple Pay is very high. So folks already have their credit cards with Apple. They don't have to get their credit card out and fill out a credit card form. It's those types of things that have led to real benefits. And then we have sort of the test discounting a little bit and found that, that was not a big lever. So it's really about optimizing the flow more than anything. And then the court cases that we're watching. Really, there's 2, Apple versus Epic and Google versus Epic. They're both going through the courts. Apple versus Epic is in appeal, and we're waiting to see how those both play out, ultimately see where the fee structures land. Google did recently roll out a new fee structure, but people should note that the U.S. component of that is still subject to court ruling. The judge has essentially said that he doesn't agree with it. It's not fair enough to developers. So we're watching how that all plays out. We expect probably a ruling on the Apple. It's hard to tell, but our lawyers internally think within the next few quarters, likely this year. In the meantime, what we're doing is just maximizing all payment savings to the greatest extent possible.

All right. Last one, and we'll get you on time to your flight.

That's great.

So you're hosting a CEO Connect event on June 11, the first time you've done an event like this. So what type of sneak peek could you tell us in terms of what investors should expect around that event?

Yes, this is -- it's going to be a really interesting one, and it goes back to a lot of questions I get around the consumer, around Gen Z. So it's all about Gen Z insights. It's us sharing all the research that we have on what Gen Z is looking for, what their dating habits are, what they're looking for in a dating app. And it also goes into Gen Alpha too where we're getting some early reads on how Gen Alpha is similar or different than Gen Z. So you're going to get to hear from our data scientists, our researchers, our product experts all around Gen Z and Gen Alpha and dating.

Great. One question, okay.

Question is just -- okay. Great. Thanks. With regard to competitive landscape, Bumble has had their issues. They've made some changes. I think they just got rid of the Swipe, kind of going...

They're saying they're going to get rid of the Swipe, yes.

Yes. And so I'm just wondering if you think 3 to 5 years from now, how you would expect the competitive landscape to look different vis-a-vis 3 to 5 years ago? Is it going to be -- is Bumble -- is it still going to be sort of you and Bumble? Is it going to be AI-enabled different players with a different model? Is Facebook going to kind of encroach on someone's turf? How do you see that all playing out?

Look, I mean, we'll have to see. I think Tinder is still by far the largest dating brand in the world by a multiple. And I think it has real staying power if we can improve the product experience. And so what we're doing is we're not throwing out the swipe. We're evolving it. I think Double Date is a great example where it just totally transformed the experience from swiping to matching with pairs. Going out in a double date is exactly what Gen Z wants. It's safer. It's more fun. It's lower pressure. We've got new modes like Astrology that is totally in the Gen Z zeitgeist. So I feel like Tinder is going to continue to evolve. I think another space I see as fertile ground for the next -- forget 3 to 5 years, the next few years is really a focus on IRL. So I think apps are going to start to blend in Tinder, in particular, online versus off-line. The line is going to get blurred. And Tinder is going to become a way you can -- a tech platform or conduit for meeting people in your life. So less about swiping and chatting and more about those IRL connections. That's what Gen Z wants. We've got a pilot in L.A. We've done 30 events there. We're doing it in a way that's scalable and doesn't affect our margin profile. We don't want to be an events company. We want to be a technology company. But you -- these events are already out there. What the events need -- are the singles, and what we need as a means for people to connect in real life. And I see a huge opportunity there. I think that's going to be a big part of the road map for Tinder going forward. So that's 1 area, I would say, is IRL. The other area is, I do think AI will play a role. I think Overtone is a great bet. Can someone figure out a truly AI-native approach? It will likely be more of a higher intent use case. Think of it as like more like a digital matchmaker of sorts. And I do think there's a place for that in the category. And in terms of the competitive landscape, look, liquidity still matters, right? AI has made it incredibly easy. We got 20 seconds left, incredibly easy to build an app. You or I can go build an app tomorrow. But you or I cannot gain liquidity in a double-sided network effect ecosystem. We cannot -- AI cannot beat the cold start problem. And so I think that gives tremendous competitive advantage to brands like Tinder and Hinge that have built true scale, and to a certain extent, Bumble. And in terms of Bumble itself, like we're in a way rooting for Bumble to win because Bumble is part of the category as a whole too, and it's a category challenge. So Bumble levels up, does better for users, so does Tinder. Hinge keeps going and innovating, then the category as a whole grows. There's much more room to grow the pie than there is to divide the slices. Just remember, category adoption is 30% or so in developed markets and 10%, I think it's like 6% or 7% in developing markets. So really, the long-term play is grow the pie, we're less worried about how the competitive landscape shakes out.

I think we end up with -- thank you.

Thank you so much.