Ladies and gentlemen, thank you for standing by, and welcome to Coursera's First Quarter 2024 Earnings Call. [Operator Instructions] and please be advised that this call is being recorded. [Operator Instructions]. I'd like to turn it over to Cam Carey, Head of Investor Relations. Mr. Carey, you may begin.

Hi, everyone, and thank you for joining our Q1 2024 earnings conference call. With me today is Jeff Maggioncalda, Coursera's Chief Executive Officer; and Ken Hahn, our Chief Financial Officer. Following their prepared remarks, we will open the call for questions. Our press release, including financial tables was issued after market close and is posted on our Investor Relations website located at investor.coursera.com where this call is being simultaneously webcast and where versions of our prepared remarks and supplemental slides are available. During this call, we will present both GAAP and non-GAAP financial measures. A reconciliation of non-GAAP measures to the most directly comparable GAAP measure can be found in today's press release and supplemental presentation, which are distributed and available to the public through our Investor Relations website. Please note, all growth percentages refer to year-over-year change unless otherwise specified. Additionally, all statements made during this call relating to future results and events are forward-looking statements based on current expectations and beliefs. These forward-looking statements include, but are not limited to, statements regarding the potential impact trends affecting our industry and business, and factors affecting the same, the anticipated benefits and impact of our strategic assets and platform advantages, including our AI and machine learning initiatives and offerings, our ecosystem platform content and partner relationships, our anticipated plans and the anticipated advantages and benefits thereof, our strategy and priorities, our share repurchase program and cash and capital allocation and our vision, business model, mission, opportunities, outlook, financial business and otherwise and future intentions. Actual results and events could differ materially from those expressed or implied in these forward-looking statements due to a number of risks and uncertainties, including those discussed in our press release, SEC filings and supplemental materials. These forward-looking statements are not guarantees of future performance or plans, and investors should not place undue reliance on them. We assume no obligation to update our forward-looking statements, except as required by law. And with that, I'd like to turn it over to Jeff.

Thanks, Cam, and welcome, everyone. It's great to be with you all. In Q1, we continue to make progress across our business as we navigate a dynamic environment and evolving educational landscape. We welcome nearly 7 million new learners. We expanded our catalog offerings with new professional certificates, pathway degrees and generative AI content. We made significant progress on getting our latest AI-powered product innovations into the hands of our learners and enterprise customers. And we grew revenue 15% over the prior year period, while generating more than $8 million in adjusted EBITDA and $18 million of free cash flow. Our first quarter revenue was lighter than we anticipated. We are seeing softness in our North American paid learners, which have an outsized impact on our in-quarter and future performance given their higher monetization rates. Ken will discuss the dynamics in more detail as well as the implications on our outlook for the year. That said, our focus on branded, job rolling credentials continues to resonate with the individuals and institutions that we serve. And I continue to see signs that the ecosystem that we've assembled is learners, universities, educators, businesses and governments all around the world, have put Coursera in the right position to help solve the fast-changing demands of the labor market and education system. So let me start with an update on how we see these global trends evolving. The first trend is digital transformation. For many years, the combined forces of technology and globalization have accelerated the transformation of nearly every institution in our society. It's been more than a year since the remarkable capabilities of generative AI mesmerized the world. Early concerns about this technology reflected challenges the world experienced in 2020 including its potential to disrupt global economies and educational systems, displace millions of…

Thank you, Jeff, and good afternoon, everyone. As Jeff mentioned earlier, we are pleased with our strategic progress on a number of key initiatives. But I want to begin my remarks today by making it clear that we are not satisfied with our revenue growth. In particular, our revised outlook for 2024 revenue. In the first quarter, we generated total revenue of $169.1 million, which was up 15% from a year ago. Growth was driven by double-digit increases across all 3 of our segments, but we underperformed in consumer, more on that momentarily. Please note that for the remainder of the call, as I review our business performance and outlook, I'll discuss our non-GAAP financial measures unless otherwise noted. For the first quarter, gross profit was $91.7 million and a 54% gross margin, in line with the prior year period. Total operating expense was $88.2 million or 52% of revenue, down 10 percentage points from the prior year period. For the individual line item components of OpEx, sales and marketing expense represented 29% of total revenue, down 1 point. Research and development expense was 13% of revenue, down 7 points. And general and administrative expense was 10% of revenue, down 2 points. I remain pleased by our ability to balance our growth initiatives and long-term investments while demonstrating the leverage inherent in our model as our platform scales. Q1 net income was $11.9 million or 7% of revenue, and adjusted EBITDA was $8.3 million or 4.9% of revenue. As a reminder, we do not optimize EBITDA performance in any single quarter. Rather, we set an annual EBITDA margin target and work within that plan based on the trajectory of the business. Most importantly, Q1's bottom line performance marked a strong start to the year as we continue to manage to…

Thanks, Ken. I'd like to close today with a special use case. Earlier this month, I joined the World Bank in D.C. to speak to Ministers of finance from emerging economies. 20 of the 30 countries represented in that room had partnered with Coursera during our COVID response initiatives many through our work with the Commonwealth of Learning. One of the countries served by this program was Guyana, whose Ministry of Education used Coursera to train more than 36,000 Guyanese citizens with more than 190,000 course enrollments during COVID. Last month, I joined the President of Guyana and many of his ministers to launch a national training initiative that offers every Guyanese citizen and public sector employee access to Coursera. The national program will be delivered through various ministries across the country with customized learning programs for each sector. For example, the office of the Prime Minister is using Coursera to skill public servants in digital media, communication and journalism. The Ministry of Human Services and Self-Security is using Coursera to train over 4,000 women in entrepreneurship, digital finance and resilient skills. The Ministry of Health assembled at the event 900 nursing students and 800 nursing assistant students who are part of the nursing school program now moving online on Coursera. And the Ministry of Tourism, industry and commerce is working with Coursera to enhance their staff skills in digital marketing, communication and data analytics for hospitality and travel. Similar to our partnership with the University of Texas System, this use case with Guyana provides a powerful example for other institutions who are looking to provide high-quality education and skills training opportunities to large populations that otherwise wouldn't have access to help citizens and local businesses unlock their full economic potential to address skilled labor shortages in high-demand industries and to diversify and drive economic growth in a fast-changing global market. Generative AI acts as both a disruptor and an enabler. Whether it widens or narrows the opportunity gap hinges on our ability to make education and skilling equally accessible on a global scale. Businesses, governments and academic institutions, we'll have to work together to mitigate the human cost of AI disruption and create more equal opportunities for everyone in a world of accelerating change. We are proud of Coursera's role and especially our partners we work with who are turning the threat into an opportunity because of education. Now let's open up the call for questions. Thank you.

[Operator Instructions]. We will now take our first question from the line of Rishi Jaluria with RBC.

Wonderful. Maybe I want to start with the softness in consumer. And there's 2 pieces there, right? Number 1 is, when did you realize this was going to become an issue? And you said that there are certain steps you can take in terms of accelerating certain content partners to help mitigate that. Maybe has that already started? What has been traction from there? And the piece number 2, within consumer. I guess I'm still a little struggling to understand why timing of one content release, which is now live, is causing such a dramatic impact on the overall revenue trajectory. Maybe help me understand built this piece, and then I've got a quick follow-up.

Rishi, this is Jeff. Yes, thanks for the question. I mean, basically, we're off to a pretty slow start on consumer in 2024, obviously. To the question of when we sort of moved, I mean a lot of it is predicated on the launch of a new piece of content. To your question, that content has launched and sort of relevant to the launch was also paid media that went along with it, particularly in North America. So the delay of the content, delayed some of the spend, we saw some of the traffic lower than we had anticipated because of that delay. And even though we're seeing good results so far, those lost months will not be recovered in terms of revenue. So that's going to be something that follows us at least through the rest of the year. Ken, maybe you could talk a little bit about the question on sort of the impact of the slower start in 2024 to our Q2 and 2024 on consumer.

Yes. And I guess first and foremost, what we highlighted was the underperformance in North America broadly with lower volume and conversion of paid learners. The delay in the content launch compounded it and certainly resulted in some of the underperformance conversion, but it wasn't the primary point. And Jeff, to your commentary, it was really about the lower marketing spend. So as it relates to that content launch. So lower marketing spend to get lower conversion rate because that is highly qualified traffic. And so we saw way overall on the results, but it wasn't the primary reason, but a contributing factor.

And then, Rishi, in terms of mitigation, I mean, one is, obviously, get the content launched and put good money behind it. Another is to say, "All right, what kind of content is getting a lot of traction right now?" and probably not surprisingly, it's AI content. The idea that people want new AI content, both for the builders who are building these models and obviously, major compensation packages going out there from companies looking to find the builders. But also the users, people who need to learn how to use this stuff, we see broad appetite 4x what we saw last year in terms of people taking AI-related content. And of course, the population of users of AI is much bigger than the population of builders of AI. And so accelerating content launches that had to do with AI and also upgrading existing content so that it has sort of this generative AI module that says, here's how you do this job in a world of generative AI are a few of the things that we think are going to be promising throughout the rest of the year. And so we do see some steps to translate the demand for generative AI into content launches, at least partially make up for the slow start that we had in 2024.

All right. Got it. That's really helpful. And then just a quick follow-up. So the NRR for paid enterprise, dropped to 94%. And that's in spite of the strength that you saw in government and for campus. So that would imply that the NRR specifically for Coursera for Business was a lot softer than that. Maybe, again, what is causing that? Is that -- layoff still happening in the customer base? Is it just deemphasizing learning and development within those customers? And more importantly, what steps can you take outside of seeing an improving macro to see that NRR get back above 100%?

So Rishi, so the -- we have seen better success in both C4C and C4G this year on overall basis. And C4B, you're right, has been where we've consistently underperformed. We've talked a little bit about product market fit and what we're emphasizing versus the other 2 verticals, C4C and C4G. However, for this quarter as it relates to NRR, the pain was spread a little bit. It wasn't just C4B, which is where we've seen some weakness over the last year or so we've been talking about that pretty actively. But it was also a tougher quarter on C4G, which is what overall was the change that caused it to drop to lower than it was last quarter to the 94%. So again, weakness in C4G this particular quarter, which were 2 specific renewals and without going into detail on them, it's not something we expect to see on an ongoing basis. The fit tends to be strong in C4G, and we get pretty good renewal rates.

And then, Rishi, on the question of how do we -- how can we manage NRR and try to get it up. I mean, clearly, great content that people want, generative AI Academy for companies who are starting that process of reskilling their employees, which is going to be kind of a top-down CEO imperative when they really get into action. But part of it, too, is use cases around what kinds of learners are taking these. I mean we actually see pretty high NRRs on this use case of students taking Coursera as part of their degree programs. I mean, students are engaged. They're there to learn. They're getting credit. They're getting it in micro-credential, that's looking pretty good. So part of it, too, is finding use cases where we know that there's good ability to get to the learners and the learners are interested in advancing their career through learning certain job skills. And we think a lot of what we're seeing continues to be fairly early market traction where, in not every case, do we have exactly the right product market fit when it comes to the learners ultimately taking some of these courses, particularly in the government segment, where Ken mentioned we had a little bit of a couple of exceptional cases where there wasn't a great fit with the learners there.

[indiscernible] COVID days, right?

Right.

Next question comes from the line of Ryan MacDonald with Needham.

If we separate out the sort of delayed course launch and the impact on consumer, you talked obviously about broader softness and lower conversion rates. As you look into the problem, is there specific reasons that or things that you're seeing commonalities of maybe what's driving that lower conversion rate? And I'd love to hear color on that. But then as we think about the rest of the years, is there ways to try to boost that conversion through discounting or any sort of other mitigating factors that you might be able to combat sort of that lower conversion rate domestically?

Yes. Thanks, Ryan. So I think with respect to conversion rates, and it is more sensitive in the North America region. A lot of what stimulates higher conversion rates is more recent launches. I mean, generally, the people who come for a more recent title are more interested buyers. And in North America, they have higher disposable income to actually pay for it. We have seen some of the highest conversion rates are in the AI content. So I think kind of an obvious mitigating strategy is to really continue to lean into generative AI, and say, not only launch new stuff for the builders and the users of generative AI, but upgrade and relaunch existing content because every indication I've seen and others could be their own judge, is that this generative AI will have a huge impact on the way people do their jobs. They're going to need to learn new skill to be, you name it, a PR comms person or a financial analyst or a supply chain manager or a UX designer, we think there's a very broad opportunity to really refresh the content, including the longer form to appeal to this very apparently strong demand that we're seeing from learners around generative AI. So that's, I think, one vector of mitigation that we are certainly going to be pushing on.

Helpful. Maybe just as a follow-up, switching to the degree segment. Obviously, there's been sort of a lot of turmoil or change within sort of the OPM market generally over the last 6 months in terms of M&A and struggles from other vendors. Are you seeing any opportunities to bring in new degree partners as a result of that period of transition that we're going through?

Yes, Ryan, we're certainly seeing inbounds. But many of the programs that are out there, they were designed for a time that no longer exists. They were designed in a time where an online degree cost the same amount as an on-campus degree, which could be $100,000 or close to that, without economics could justify spending $25,000 to acquire a learner. And those types of programs just don't really fit our model. So with the pathway degrees where we're really pushing far more affordable credit pathways, so that you could start an open content and have that count towards a credit degree. We don't see a lot of those types of program designs in the portfolios of some of these traditional OPMs who've been struggling. And so I would not expect -- we're not anticipating seeing a lot more supply of degree programs coming on the Coursera, they are essentially transplants from those traditional OPM players.

Your next question comes from the line of Stephen Sheldon with William Blair.

First one on the guidance, it assumes about 7% year-over-year revenue growth in the second quarter and then a reacceleration back closer to 10% in the second half of the year. So just curious what gives you the confidence in that second half reacceleration?

Sure, Stephen. That's primarily in consumer where we expect the acceleration. If you look at the guide we gave across the different verticals. And quite a bit of that is product driven. We have a number of products coming out in the second half, particularly international payments is something we've talked about internationally in pricing. And so we've had and we originally had put in our initial plan at the beginning of the year. Weighted average improvements that we expected to see from operating better with those products. So that's the reason for the uptick in the growth as we go into Q3 and Q4 versus Q2.

Yes. And some of the product innovations will basically take advantage of the translations and we want a couple of those translations, which have obviously created a broader accessibility with, as Ken said, pricing, geopricing, payments, currency and also merchandising in international markets. So we're planning to see higher NPL -- new paid learners and better conversion in international markets. And we also do have a number of titles that we think we'll be launching in the second half of the year that we're pretty excited about and follow the same basic pattern of other branded professional certificate types of titles that have seen good success and good demand on platform.

Got it. That's helpful. And then what are you seeing in terms of interest levels from perspective, of Coursera for campus customers to do something similar to what you've already launched with the University of Texas. Has interest there picked up, I guess, this year has progressed?

Yes. I think that as we continue to show other use cases, that could be happening within a single campus, well, actually, within a single school on a campus, the total campus, a system of campuses or even a national system like in Kazakhstan, we are definitely putting together more and more use cases. Obviously, universities are not institutions who are known for their agility. But I think just the inevitable unyielding force of change that is saying, I mean, not only are working adults looking for more flexibility and affordability with the degree programs. The curriculum that they're looking for is different. I mean, it's turning out to be a lot more digital and especially not with generative AI, you can imagine how difficult it is for the universities to keep up with the curricular offering. And then you layer on top of that the need to actually transform the organization, just the productivity, the operating leverage, the way that you run a university college or other -- vocational school is also changing. So we are seeing a recognition that is fairly incontrovertible. This just is like, yes, this is going to be a very big change. We're going to need some help. And we think at Coursera, that Coursera is an ideal partner and pretty unique out there. There's just not that many other players who can offer electives for the students that can count for credit and a Generative AI Academy to help transform the organization.

Next question comes from the line of Josh Baer with Morgan Stanley.

Great. I wanted to come back to the Enterprise segment. First, I guess, it was one of the strongest net enterprise paid account adds sequentially in a while. So I was wondering what drove that. And then with that in mind, is that a read-through to like stable churn overall? Or were the gross adds, particularly strong, but you -- but the account churn was part of what contributed to the lower net retention rate?

Yes. Josh, so it's a few things. One is average deal size is coming down a bit. And so companies are interested and sometimes changing out their current partners and saying, "All right, look, we'll give you a certain number of licenses to work our data scientists or to work with our marketing department or to work in our leadership program. And if that goes well, we'll expand it." And so part of it is kind of reshuffling a portion of an existing budget over to Coursera to say, "we want to try your approach to this whole thing." And then we also see a world where the budgets are stabilizing a bit. I do, quarterly, I do these events where we have -- I have discussion forums with Coursera for Business learning and development people in the different regions of the world. And I've been asking, "is your learning development budget getting bigger, getting smaller or holding the same?" And there's a, at least from somewhat anecdotal pooling, they have been saying, "Hey, we're getting ready. There's going to be a big need for rescaling due to generative AI. We're starting to see a change in the budget." And so I think we're also picking up a little bit of a stabilization. And ideally, I'm optimistic that they're going to realize and when you look at the survey that I mentioned in the script, survey after survey is showing that even though almost every CEO and executive says, major organizational skilling and transformation is going to be required, they also say that so far, they've almost not even started yet. Well, once they get their team in order, their strategy in order and their budgets in order, we do think that there's indications that there'll be good dollars being spent on generative AI-related upskilling and reskilling in businesses.

Got it. And then just like one follow-up on the segment guide, 10% across the board. With consumer coming in, in the high teens this quarter, like the 10% for the year, seems to make sense in the context of everything you're saying. But enterprise was -- is already at 10%, I think, and move lower sequentially. So could you talk a little bit more about that stabilization when the net retention rate is below 100%. And we just saw that revenue tick lower sequentially.

Yes. So thanks, Josh. Question B, the -- the methodology was as simple as renewing our forecast and then taking a look at the overall growth rates with each of the segments. In particular, the enterprise group. And enterprise remained at 10% as it was with the previous guidance last quarter as well as degrees at 10%, which I'm sure you noted. Enterprise, again, the NRR, there is a larger shortfall than we would have expected for some onetime contracts to onetime contracts on the government side. We do not expect those to recur. And so we think you will see the NRR bounce back, which is the biggest lever on the revenue. So that would explain the deceleration and then moving forward and still achieving the 10%.

Your next question comes from the line of Brian Peterson with Raymond James.

This is for Jeff or Ken, but would love to understand your latest thoughts on payback and your sales and marketing investments. And is there any change in how you guys are thinking about sales and marketing with some of the management changes you guys made, would love to get more color there.

Yes. This is Jeff. So, I think there's obviously a difference between the consumer marketing dynamic and then the enterprise where we've got the direct sales force. On the consumer side, we still continue to see pretty good returns on average spend on consumer, notably with these professional certificates and especially for the newer ones. So we're feeling pretty good that there's demand out there. And if we market these things, we get a good return on it. We're also seeing in international markets increased return on average spend given the translations, we think that the translations have kind of opened up some opportunities to deploy paid media outside of the U.S., and we're feeling pretty good about that. And then on the enterprise sales team, I think, frankly, a lot of -- I've been out there in a lot of these deals. Companies are -- they know that this generative AI training is a big deal. They're trying to get their act together. They're trying to figure out what is their playbook. How are they going to actually scale? What groups do they go with first? Right now, they're still focused a bit on the builder saying, do we have people who understand this stuff. But I think creating some shape for those larger generative AI Academy deals will help. And so right now, the deal sizes have been on the smaller side. I'm optimistic that if they start doing larger scale training of generative AI, those mandates might expand, and that will create a better return on our enterprise sales and marketing spend. By the way, I'd also say on Coursera for Campus. We are really focused more and more and more on the for students for credit use cases where the students get the best value and the institution gets the best value. It does take faculty a little time to warm up to this idea. I mean usually, the University President is quite interested in it. They've got to make the decisions and sort of enact the policies to say, yes, we're going to offer these industry micro-credentials as career electives for credit. That takes a bit of time. So there's a little bit of market development that we're doing with our sales and marketing team and Coursera for Campus. We, frankly, I like that investment. I cannot see a way that higher education is able to respond to the changes without something that looks a lot like what we're offering with these industry micro credentials that can count as credit towards [indiscernible] degrees.

Understood. Thanks for the color there, Jeff. I know you mentioned a lot about AI content. I'd love to understand how do you think about AI in terms of the professional certificates of the consumer side? And where is the ecosystem of partners in terms of really enabling that AI content for those really higher price point consumer courses?

Yes. I think that the -- I suspect that the professional certificates with generative AI are going to look a lot like the certificate programs that all the cloud players have been doing for the last 5 to 10 years. The basic idea is in order to promote their platform as a leading platform, whether that's AWS, or whether that's Azure or whether that's Google Cloud or IBM, they're going to want to essentially not just merchandise their platform, but train people and certify people on their platform. We'll see, but in the past, the platforms have only certified the builders on those platforms. I can see a world and we're getting indications that a lot of the platform providers want to also create certification programs for the users of those platforms, of which, of course, there are orders of magnitude more users of the platforms than builders of the platform. So if things go the way that I think that they might, there could be a next wave of certification programs but on potentially a much larger scale where you're certifying not just the builders, but the users.

Your next question comes from the line of [ Devin Alp ] with KeyBanc Capital Markets.

Great. Not sure if you already mentioned this, but could you share the specifics on what drove the delay in that content launch from your education partner? And are there any guardrails or initiatives in place now to prevent that from happening again down the road?

Yes. This is, Gavin, this is Jeff. So there -- we only have a certain amount of visibility into the production processes for different partners. I would say that one of the things that Ken talked about last quarter, which we continue to lean into for a number of reasons, is us assisting our partners in actually building out this content. A, we've got a pretty good idea of what the success factors are to design and create a good piece of content; and b, more and more industry partners here to want to work with Coursera to put their branded content on Coursera, but they're not instructional designers, so they're looking for us to help. The benefits to us are a few. Number one, we can have a basically more exclusivity on the content. Number 2 is if we help to build it, and we're putting more resources towards building it, we get a better share of the economics. And then number 3, to your question, we get better visibility into that pipeline and actual management of the production process. So I would say that for a number of reasons, including these Generative AI tools making our productivity much, much better and creating high-quality content at lower prices we are basically providing more and more assistance to our partners in creating these certificates, and that will give us both more control and more visibility.

Great. That's helpful. And then just staying on consumer, with the lower volume and conversion of paid learners you called out in North America, does that softness have any potential impact to, I guess, degree segment in terms of the funnel and filling those cohorts and degree pathways in the near term?

It might. I don't know that the numbers are significant enough to materially affect that. What we find is that when we compare the right college degree with the right pathway policy, so you can get credit to the right segment of learners and a professional certificate, you can unlock much higher conversion rates. And so I think it's a bit more about getting the right pairing of certificate degree that does it. Our NRLs, our new registered learners, are also looking pretty good. So we feel at the top of the funnel, we brought in another 70 million globally, although you're right, in North America, it's a little bit soft, we are feeling pretty good about the overall site visits and new registered learners. So I'm not too worried about that. I don't think that will become a material impact on our ability to fill degree student cohort. I think it's much more about unlocking these pathways and really improving the conversion rate because of that unlock between the professional certificate and the pathway degree. Ken, anything you'd add to that?

No. Said the exact same thing. Thank you.

Let's take 1 final question. Your final question comes from the line of Brett Knoblauch with Cantor Fitzgerald.

Maybe just on comparing the LTV of your North American learners to that of the international learners. How big of a delta is it between them? And then do you think the new payment and currency functionality will help narrow that gap?

Yes. Great question, Brett. I will say there's certainly a difference. Part of it is the price for when we do geo-pricing part of it is the retention, which is typically longer. So retention times price gives you the LTV. Ken, can you give any general way to think about it? Obviously, it's not something we typically disclose.

No, we haven't -- it's a multiple is the answer. It's a significant difference between North America and Western and all else. Not surprisingly, it's about per capita GDP, especially in APAC. So we haven't given broad guidance, but it's a multiple. It's a significant difference. As far as new product, enabling that more or creating a change in that, we do think we'll be able to realize more value. If you think about it fundamentally, what we're doing around the translation is we're making it accessible. So we think that should drive both volume and over time, the ability to -- for pricing. And payment systems as well, making -- enabling the consumer to transact where it's difficult for them to do today, we think will give us both pricing flexibility in the future, but more important, we'll produce more volume. So partially on the LTV, you'll see increases, but a lot of it is a function, again, per capita GDP. And there's more opportunity for us to close that gap from a value-add standpoint internationally just because we haven't done these things historically. So you'll see a little bit of both we expect.

Perfect. And then if I could just have 1 more. On the consumer side, I guess, how much of the revised consumer guidance is attributed to the delayed content versus anything else like could it be macro starting to impact consumers' budgets and with inflation continuing to be high, is that affecting maybe the demand funnel?

Yes. It is both, Brett, to your question. So there was a general slowdown in the conversion. We can see some of that as a result of the delayed content launch and particularly the effect of less marketing dollars going against that content launches. Against a little bit to your previous question, our highest performing, highest value region from a consumer standpoint. And so it is both and we haven't broken out the attribution that's hard to understand exactly between the 2, but it's an element of both. But it has -- the slowdown at the beginning of the year around the marketing affects conversion and run rate for the rest of the year, of course, because it's a building both conversion -- the programs build over the first couple of months. And so by delaying that, it's -- we lose that for the year.

That wraps today's Q&A. A replay of this webcast will be available on our Investor Relations website, along with the transcript in the next few hours. We appreciate you joining us.