The New York Times Company (NYT) Q1 2021 Earnings Report, Transcript and Summary

Hello and welcome to The New York Times Company's First Quarter 2021 Earnings Conference Call. All participants will be in listen-only mode. After today's presentation, there will be an opportunity to ask questions. Please note, today's event is being recorded. I would now like to turn the conference over to Harlan Toplitzky, Vice President of Investor Relations. Mr. Toplitzky, please go ahead.

Thank you and welcome to The New York Times Company's First Quarter 2021 Earnings Conference Call. On the call today, we have Meredith Kopit Levien, President and Chief Executive Officer; and Roland Caputo, Executive Vice President and Chief Financial Officer.

Thanks, Harlan, and good morning. The Times finished the first quarter with more than 7.8 million paid subscriptions across our digital and print products, more than 100 million registered users, and an average weekly audience of 76 million readers. That foundation, plus our unmatched journalistic breadth and a market of at least 100 million people, who are expected to pay for English-language journalism, grounds our conviction that we can substantially and profitably scale paid subscriptions over time. Our strong financial results in the first quarter demonstrate the success of our strategy and the promise of our large and growing digital subscription base. We recorded adjusted operating profit of $68 million, an improvement of more than 50% compared to the first quarter of 2020. Digital subscription revenue increased 38% in the quarter and we added 301,000 net new digital subscriptions across News, Cooking, Games and Audm. Our first quarter results also reflect a real improvement in digital advertising. That improvement was driven by a brisk market and our work last year on refining the competitiveness of our offering and the margin profile of our business. Digital advertising growth over Q1 in both 2020 and 2019 demonstrates the advantage of our subscription-first strategy to our advertising business. In our last earnings call, I noted that fluctuations in the news cycle can lead to considerable variability in net digital subscription additions from quarter to quarter. In February and March, our audiences declined from their historic highs last year, and we saw fewer net subscription additions in the latter part of the quarter. We expect moderated growth to continue through the second quarter, traditionally our softest of the year. With lower forecasted second quarter performance, we now expect annual total net subscription additions to be in the range of our 2019 performance, which prior to 2020 was our best year for net additions.

Thank you, Meredith, and good morning, everyone. Adjusted diluted earnings per share was $0.26 in the quarter, $0.09 higher than the prior year. We reported adjusted operating profit of approximately $68 million, higher than the same period in 2020 by $24 million. We added $167,000 net new subscriptions to our core digital news product and $134,000 net new subscriptions to our standalone digital products, for a total of $301,000 net new digital-only subscriptions. As of the end of the quarter, we had nearly 900,000 Games subscriptions and slightly more than 800,000 Cooking subscriptions. The international share of total news subscriptions remained at 18% as of the end of the quarter.

Thank you. We will now begin the question-and-answer session. And this morning's first question comes from John Janedis is with Wolfe Research.

Hi. Good morning. Meredith, given your slightly moderated tone for subscribers this year, can you talk about your confidence level around the longer-term opportunity? I know you talked about 20 million or more subs over, say, longer-term. Does that still seem doable? And big picture, how tight is the growth to the new cycle?

Good morning, John, both good questions. My short answer to your first question is absolutely no change in our level of confidence in the long-term opportunity. We see no reason why we can't have a subscriber base that's 2, 3, 4 times what we have today. We remained incredibly excited about that, that long-term opportunity. On your second question, I think you're asking me the degree to which the news cycle plays a role. Just ask the second question again. I want to make sure I've heard it right.

Yeah, I mean, just based on the comments around late 1Q and 2Q, knowing it's seasonal. And obviously, there's been a lot of talk about a slowing news-cycle have an impact to subs. And so, I guess, one narrative would be, if the news-cycle slows, is there more of a longer-term discussion here that we should have around, say, the subscriber acquisition?

Yeah. So, let me say a couple of things about that. The first one is, of course, the news-cycle plays a role and has played a role in many ways. The various stories the last 5 years, which is really the stories of the last year, had been an accelerant to the model, but that I would say 2 other things are happening that are really significant. One, quarter-over-quarter, year-over-year, we are getting better at the underlying mechanics, getting better at the underlying engine, driving subscription and building habit. And as I said in my prepared remarks, we feel really pleased with where we are today from an engagement standpoint and the game habit standpoint versus 2019. So, we feel very confident that there is still wide interest in news. I'll also say, I'll say 2 more things. I don't think the world is getting any less interesting. I don't think it's getting any less complex. And we fully expect, even as the tide goes out on one set of stories that it always comes back in from another set. So we are not worried that there won't be high demand for news. Even as the storylines change, and I talked in my prepared remarks about the enormous investments we've made in our newsroom. And we are prepared to meet some of that investment. We are prepared to meet that demand, wherever the story goes. I'll say, we've also did quite a bit of investment into meeting news needs in new and different ways. I mentioned real-time coverage, live news, that's a need, where I'd say, people didn't come to The Times first before, I'd say before 2019 in a really big way. It's something that's happening, unfolding in real-time. You might have turned on cable news or gone to Twitter, and we are getting much better, still lots of room to do but getting much better, meeting that need today. We are getting much better at meeting the need for the daily habit of showing up in your inbox with a morning newsletter and newsletters beyond that. So, I'm really confident that we're going to keep meeting the need for understanding the news. And we are going to add and innovate how we do that. And that news will continue to be the central engine of the model and a real creator of demand that helps us sell the whole bundle.

Got it, thanks. And maybe, can you talk about media expenses going forward? Do they start to pick up significantly to drive or reaccelerate subscriber growth? And can you give any more color on churn?

Sure. I'll answer the first question, and the second one, Roland may have color to add to that. On media, I'll just remind you that the vast majority of our subs come in through organic means, that actually goes a little bit to the point I made before. Our news product is a huge kind of self-propelling engine of sub demand. So, we are still getting those to our subs, not through paid media. In periods of lower news cycle or less timing. And we also end up buying less media, because the rate of return isn't as good. But the overarching engine is much, much more reliant on organic means. I'd say, on churn, we are broadly very happy with churn. When we compare ourselves to other content-based digital media companies, we think our churn - our overall retention and monthly churn are at a world-class level versus last year, I'd say that the newest cohort, a huge surge in new subs last year and the newest cohort, subs that came in last year, are retaining slightly less well than in the past, but I'd say not in a troubling way. And we're broadly very happy with what we're seeing on churn. But, Roland, I don't know if you want to add to either of those points.

No, I actually really think you hit all the major points, nothing to add on that one.

Thanks a lot.

Thank you.

Thank you. And the next question comes from Alexia Quadrani from J.P. Morgan.

Thank you very much. Just a couple of questions. First, if you could maybe elaborate a little bit on what the demos look like in the new subscribers in Q1, if it's any different than in the past, sort of mix of U.S., international, that sort of color? And then, secondly, Meredith, you have - I think the outlook for digital advertising has changed on and off throughout the year as the world has evolved and the model has changed a bit. I'm curious, I know you mentioned it's being very, very healthy outlook for digital advertising for the remainder of the year. I'm curious if your long-term view of digital advertising has changed at all or maybe you can update us on how you see that, how the outlook there is?

Sure. I'm happy to talk about both of those. Thanks, Alexia. On demos, in general, I think I give a version of this answer for every quarter - in general, in very high news periods, where we're bringing a lot of people in, the audience tends to be getting a little bit younger and a little bit more varied geographically, and that holds up both domestically and internationally. So, it's we've been sort of steadily tracking toward more and more of the new subscribers coming in under 40, and more from parts of the country and the world, where we've not been hugely penetrated. In quarters, where there is sort of less - kind of high news demand, it tends to go the other way. But I'd say, very broadly, The Times is continuing to bring in younger subscribers and subscribers from more varied places. And there is nothing that happened in the first quarter that changed that trajectory. And I think I've also said in the past, we don't ask a ton of questions demographically of our news subscribers. We do more of that over time than the beginning. So our data isn't great here. But there's generally - and they're generally more varied as we grow the base. On advertising, I'll say a few things. Our ad team did very, very good work last year in a really difficult ad market and margin profile of the business. And I'll be specific about that. We exited our less profitable marketing services business. We've had a thesis, gosh, 5, 6 years ago, that can be really additive in our ad business. And in fact, what we've learned is that the demand is really highest for our new products, including fundamentally a higher-margin endeavor. Demand is getting stronger and stronger for our first-party data products, which confirm registration-based model, we're now making faster-than-expected progress on. And the creative work that we - we still do a bunch of creative work. But we did that previously through fairly cumbersome lower-margin marketing services business. And so, now we're able to do that work in a way that is nearly always dependent to media. And therefore, at deal level, we like the margin better. So that's all a long-winded way of saying, we like the profile of the ad business, the digital ad business better than we have in some time. I still would say, no question, it's the secondary endeavor. And the main idea here remains scaling our direct-to-consumer digital subscription business. I do think now we've gotten those 2 businesses. The real thing we pulled off last year is we have those 2 businesses running like entirely on the same high-octane gas, which is registered logged-in users, whose data we can use in privacy in more ways, and we like that a lot. And I think overall, it does mean a better business, it is still not the main idea, not the main growth driver.

And if I could just squeeze in one more, just circling back to your comments earlier about churn, just tweaking up just slightly? I guess, it's understandable, given the huge number of subscribers that came on last year. I'm just curious if it changes at all or how do you revisit the algorithm in terms of the price hikes coming off of the promotional rates, given that tweak in churn or not necessarily?

Yeah, I'll let - I'll say 2 things and Roland should add more color here. We're very comfortable with where churn is. And I'll just say again that it's been one of the biggest accomplishments over the last 4, 5 years. We see dramatically increased retention and reduced churn. And we're very comfortable with everything we see in the model and where it is exactly, right? You're bringing in giant surge of new people in a single year. They are slightly less qualifying, so that the newest cohort is slightly less sticky, but only slightly. And churn broadly is still a very, very good story. Our algorithms are getting a lot better. That's the point. You train the models and it may get smarter. And so, we also like what we're seeing. And Roland touched on this in his prepared remarks, our pricing strategy and our ability to use algorithms to determine who goes to full price and who doesn't is working very, very well. Roland, I don't know if you want to add anymore color to that.

Sure. Hi, Alexia. We are still really comfortable with our pricing strategy and the way we're going about it with the modeling. As a matter of fact, we actually are testing a second intermediate price, which sits somewhere between the intermediate step-up price we've been using and full-price to try to take a little bit even more advantage of the demand curve. But we haven't increased or decreased the number of folks. The model hasn't increased or decreased the number of folks that it is sending to full price, so still comfortable with the model, no problems there.

Thank you very much.

Thanks, Alexia.

Thank you. And the next question comes from Kannan Venkateshwar with Barclays.

Thank you. Roland, just to follow up on the previous comment you made on the intermediate price, is there any difference in the kind of subscription that people get for the intermediate price, or is it just some kind of a bridge till you step people up to full-price? And then, broadly, when you think about price increases over the course of this year, I guess, there is a cohort that already has, which came in 2018 during the promotional phase, which has either seen a price increase already or will see a price increase over the course of the year, which is the second time their price would be raised, I guess, in the last 2 years. So, could you talk about the churn of that particular cohort, and how that's behaving and retaining? And lastly, on the advertising side of - sorry, I'll follow up with advertising after that.

Okay. So, good questions. And actually, there is not a differentiated experience for subs across this pricing range. What we're really trying to do is not push price on someone who is not apt to accept the price increase as much as someone else would, so we're trying to divide the population in that way. But there is no difference in the subscription experience that someone going all the way to full price versus someone going to a midpoint price receives. Your second question was sort of about that second moment, that second price increase. So our experience to date with that is we're very comfortable with that. It is another test of that person's elasticity of demand. So we do see a bit of drop-off there, but we're perfectly comfortable with the level that we're seeing there. And, of course, we will take action to lean a little bit into trying to save the person and keep them with the brand than pushing the price, because we're still more interested in scaling than monetization at this point.

Got it.

Hey, Roland, the only thing is - Kannan, the only thing I would add to that is that we also, I'd say, and I think we take a conversion of this, but we get sort of steadily better. We get steadily better at retaining through engagement as well and that product proposition keeps getting better. We are investing in newsroom. We are investing in the digital product experience over time that has a positive.

Got it. And then on the advertising front, Meredith, a lot of the comments that you've made, I mean, it sounds like some of these improvements are structural. And therefore, there could be a bit more correlation between subscriber growth and advertising going forward, and we should see less volatility. Is that the right way to read your comments on advertising and should that be more linear going forward?

I think that is strategically correct with a couple of caveats. So, I'd say, having a very large - we also said today at this time, we have 100 million registered users. So, large registered user base with, whom we have direct relationship is extremely important. It's not just the subscriber base, I would say, it's the audience, the engagement, the fact that these people are fully addressable to us and we update on that. That is all very positive. That's what I mean when I say the ad business and subs business now are on the same kayak team, yeah. So, that is a yes. To the question you're asking, yes, I do think that, that bodes well for advertising. Having common to the grown up in the ad business and run our ad business, I will also say, we have a highly competitive product versus other publishers and probably any traditional media company, highly competitive. But we are still in a world where you've got Google, and Facebook, and Amazon now increasingly, you've got these giant platforms with a particular offering and scale and at a price that is unmatched. So, I'd say, we're very confident about our ability to win share. We are very confident now that in a brisk market, there is a lot of business to be won; that's going very well, if we got the right proposition. But I would just say the ad market is a finicky thing, and you've got other companies that kind of dominate the dynamics. And so, we tamper our excitement about it with that in mind. It's a demand-driven market, more than a size-driven market. We have awesome supply, probably better than anybody else is, but it's a demand-driven market.

Got it. Thank you so much, guys.

Thank you. And the next question comes from Craig Huber with Huber Research Partners.

Great. Thank you. Few questions, if I could. To start, I want to ask, if I could, about your marketing costs. They were down 21% year-over-year to $36 million. What should we expect for that dollar amount going forward for your marketing spend, and how much of that decrease by design in your marketing costs you think ended up hurting the number of new adds for digital subscriptions in the first quarter? I mean, how much of a correlation do you think there is there?

I'll give a short answer. Roland will give you a long answer. The short answer is that's by design. And that's what I said, I think when I answered the John's question, which is the biggest engine we have is organic. And when that engine - when the news cycle slows a little bit, marketing actually becomes much, much less efficient. And you don't see us spending more money because it's less efficient. As things pick up, you'll see us spending more money. Roland, you can give an even more colorful answer.

No, I think that's generally right. I mean, we don't see the cut back in marketing spend hurt our net adds at all. But, if you go back and look at the comps, we did really pull back dramatically if you go through the last year in the early half. So, I think, you'll probably see us reintroducing some marketing spend as we get into the second quarter. But, again, we gauge it, based on how well we think it's going to perform in market.

My next question, guys, for potential price increases on the digital front for the remainder of year, can you give us just a sense of how you think about phasing that in over the course of the year? How many people you're expecting that can bear with a higher price increase? I just want to get a sense for our modeling purposes.

Yeah. For modeling purposes, you think about 100,000 folks per quarter in that range, would be seeing - I'm assuming you're talking about the tenured price increase.

Yes. Yes. But, obviously, it's not a very big number. It seems 400,000 roughly for the year.

Yeah, 400,000 and change for the year.

That means you're - that's telling me that you're obviously trying to volume here, you're not trying to scare anybody away. So that model…

Well, I mean, we look at a whole set of attributes that determine when we pull the trigger on the tenured price increase. And our current outlook is that, it's a little bit over 100,000 a quarter is what we think we'll see for the rest of the year.

Roland, let me just add, that is for tenured subs.

Yes.

We also have an enormous number of people stepping up in price, which Roland had just described.

That's right. The number of people stepping up off of the big year we had last year in terms of new subscribers, you can imagine, is going to be very large. We're estimating for the full year, that's 1.5 million or 1.6 million folks will see an increase from their promotional price.

But also on the cost front, you've obviously said, you thought costs for this upcoming quarter would be up. I think you said, mid- to high-teens. I want to get to the bottom of this. In the first quarter, you said it would be up mid-single digits, it's up about 1.5%. What's the delta there, please? Why did you not spend as much on it than we originally thought when you told them?

So, it really came down to media, media spend and a bit of the hiring ramp.

Okay. I guess, my last question, Meredith. If you mentioned registered users now are about 100 million. What is the daily visitors on average you guys have now to the main website, or main app, or monthly? What's the number of that right now on your data?

That's a good question, we don't answer that one. But what I'll say is, we've more people who use New York Times every day now than we have had at any point in our history. But we've not disclosed that.

Okay. Thank you.

Higher, higher is the answer.

Okay. Thank you. And the next question comes from Thomas Yeh with Morgan Stanley.

Hi, good morning. Thanks for taking my question. Yeah, following up on the 100 million-plus registered user base, can you talk a bit about any of the strategies that you're undertaking around encouraging sign-ups through that funnel? I mean, beyond article limit that were implemented some years ago and engaging users with more content, what are some of the mechanisms you're building to convert those for users? And then, any update on your prior guidance for modest operating profit growth, given the strength in 1Q and 2Q guidance on the financials? How should we think about the cadence of some of the investments you're making? And should we expect some of those investments to accelerate in the back half? Thank you.

Sure. Let me take the second question first, which is just to say, as I said in my prepared remarks, we are building a business as an underlying model and a strategy that should propel a larger and more profitable business over time. I wouldn't regard what we said in the last call's guidance so much as direction of travel and nothing has changed in terms of our assumptions and the direction of travel, I said that in a couple of places in the prepared remarks. On your first question, I'm making sure, you guys - you know what, I just got a note from my team, you can't hear me. Give me one second and I'm going to answer the other question. I am just going to take my headsets off and pick up my phone, hold on. That's cool, holding the phone now. On your first question, I'll say a few things and you can tell me if this isn't in zone of what you're asking about. 100 million registered users now. We still do bring in a good number of new registered users every week. So there is still more activity happening on driving registrations, I'd say, the thing that's going to change now is that's less of a focus; last year's enormous audience and new cycle presented an opportunity for us to really scale that. We are incredibly focused now on getting those registered users to return. And I mentioned in my prepared remarks number of the - sort of zones of opportunity. We've got a huge ton of opportunity in e-mail. 15 million people using - opening, reading Times, e-mail every week. 5 million people who use The Daily on a very - I'm sorry, not The Daily, The Morning on a very regular basis. And so, there is a really big opportunity in e-mail to have it be a more intentional driver in the funnel from registration to subscription. And you're going to see us get just much more, more intentional and focused on that. Another place where you'll see activity with registered users is, we have very, very high engagement in our app. And once we get a registration, we're much more able to direct people to download the app and begin to use the app. So, you'll see more activity there. And then, something I mentioned in the prepared remarks, and I think I probably said in every one of these calls for some time, we know that people experiencing the breadth of our report, so more topics, more different storylines drive. It's a behavior that correlates with paying and staying. And so, you can imagine that we're working to stimulate that, now that we've got registered users and they are addressable to us. And so, I'd say, generally, you will see a shifting from driving more regis, although, I don't want to suggest we're anywhere near the top, we're not to getting registered users to return. And just the last bit there is, and maybe this is implied, but registered users convert obviously at a significantly higher rate than anonymous users, so that's why the focus there.

Yeah, that makes sense. And maybe, if I can squeeze one last in, Meredith, can you opine on the evolution of regulatory events in Australia around news on big tech platforms and any read-throughs you might have on a kind of broader opportunity there? Thank you.

Sure. I will try not to opine, although, I'll give you the state of things. And I should say 1 more thing on the last question. Roland touched on this before, but 1 of the other things that really big registration - registered user base gives us is the opportunity to keep training our algorithms. And, I think, I said a version of this in the prepared remarks, we've - or I said in prior calls, we've got a dynamic meter now. So, we are able to customize when we actually ask a registered user to subscribe and we've got machine learning that's getting better at how we do that. So, I should have said that before as well. On the platforms, I would say, we continue to watch with real interest everything that's happening in the legislative and regulatory environment. I'd say there has broadly for the last couple of years now been a shift in the direction of platform seeking and being willing to make commercial agreements in the form of licensing arrangements with publishers. We obviously have 1 of those arrangements now with Facebook for their news tab and we certainly don't rule out they're being more there. So, we're watching that with interest. We have a really clear strategy that we are in the business of scaling our direct-to-consumer digital subscriptions. And that to do that, we need direct relationships with users and we want people to experience our journalism on our destination, where we think they get the best of it, the best experience of it, our work in context. And we also believe that platforms get a lot of value from publishers, get a lot of value from the original work and news that, courses through their systems. And so, we think the compensation has to be right. So we think about all those things when we consider whether and how to work with the platform. And I'd say, I don't rule out, more to come there, but I have nothing to report.

Great. Thank you so much. Very helpful.

You're welcome.

Thank you. And the next question comes from Doug Arthur with Huber Research Partners.

Yeah, thanks. 2 questions. On the - you have a comment in the press release about a sizable and sustained investment in the journalistic engine. Meredith, is the newsroom now over 1,800? I mean, are you adding more resources there? Can you comment on that?

Sure and good morning. I've said this a few times in the past. I would the newsroom now is well over 1,700 and I'd say that number doesn't include the people we have - many of the people we have making recipes, winning puzzles, and producing Wirecutters' product reviews. So, depending on - we even think about, what do we call the newsroom? You can assume that particularly in the standalone products area that we are continuing to invest in the content, and we've said that before as far as investing in the newsroom itself. So, the work of our core news report. We have continued to invest, I'd say, thoughtfully, and we will continue to invest thoughtfully, and that investment tends to go to ensuring we can cover the biggest stories of our time in a leading way, and then also meeting more news needs. I described our budding engine for real-time coverage, live coverage, that's been an area of investment, although there, I'd say, that's as much about digital product and engineering investment, as it is about adding more journalists. We've made a big investment in audio, journalism. I named some of the podcasts that are starting to do really well from a listener standpoint and also from an advertising standpoint. So, that's how I would regard investment into the newsroom. To the extent you're asking me about, does that scale with our expectations of subs scaling? I'd say, no, not at the same pace. We really like where we are in terms of the breadth of expertise in our Newsroom and we're certainly going to keep adding to it, but not at the same level that we are scaling the subscriber base.

Okay, great. And just 1 quick follow-up. My recollection is you shut down a couple of in-house advertising operations last year. What's the impact on the year-over-year comp in revenues for 2021 in digital advertising from sort of the timing of that and the impact of...

Yeah. Roland might give a slightly more precise answer here. But I think what I said in my prepared remarks is the comp, we would have been up somewhere in the neighborhood of like 23% in digital advertising if we've not been comping against the closure of the last of our marketing services business, and also the fact that we took open market programmatic out of our apps that was the big comp as well from the first quarter.

Okay. Great. Thank you.

You're welcome.

Thank you. And the next question comes from Vasily Karasyov from Cannonball Research.

Hi, Vasily.

Good morning. Thank you. I have a question about the outlook for subscriber additions this year. So can you tell me, please, what changed and what was wrong with your model or where did you have to adjust your model from February when you were talking about adding between 2019 and 2020 number of subs, years 2019 and 2020, and now?

Yes, happy to talk about. Yes.

So - and then, what that factor is? And is it possible that that factor will change in a quarter or two, like how fluid that is?

So, let me answer the second part of the question first, which is, it is of course possible that it changes. We've said this in the past and I'll say again. There is sort of more underlying stability in the whole thing, because of the registration model. But we said there could be considerable variability from quarter to quarter, and what we saw in the first quarter was really strong January. And then, the sort of change in the news-cycle, combined with the opening up of activities that people could do sort of following a year of quarantine, combined with better weather, combined with the prevalence of vaccines that all sort of happened at once. And I'd say, it came a little faster than we were expecting, probably because the compounding effects of all of those things at once. And the reason we've suggested that it could continue into the second quarter is because the second quarter traditionally tends to be just a slower period for audience and news engagement. But, I'll say, as I said, I think an answer to an earlier question, we have seen periods, where the tide has gone out little bit on one particular storyline or few storylines and it always comes back. And we regard the underlying drivers as strong, very strong in comparison to 2019. And I'd say, we've got plenty of room ahead to optimize the model and optimize the funnel on our roadmaps now. And I'll just, and I said this before, but I don't think, it's hard to say, I mean, obviously, the news cycle plays big role in the model. And so, 2 just our work on the underlying mechanics of it, it's hard to piece those things apart. But I don't think the - it's hard not to regard this as a moment in time right now.

Thank you. And a quick follow-up, not follow up, a quick question. Why did you feel, like you want - you needed to disclose the number of registrations, the 100 million? I don't think you've talked about it before, in this level, with this level of precision. I think, you were saying that you had a very large data base.

Yeah.

So can you give us an idea?

Yeah, that's a great question. Listen, we really believe that the registration model has just improved our ability to get at those users and get them to form a habit, and ultimately pay and stay. And honestly, we were waiting for it to be a big enough number, that itself worthy of sharing. We think the fact that the number - we don't think we're done, just to be clear, the 100 million. And as I said earlier, we're still getting more registrations. But at 100 million that gives us really sizable population around whom we can intervene to stimulate return and draw people in a new way. So, we disclosed it, because we want to make the point that it's a really important part of the model, it's really working. And it's based on which we're building lots of other optimization. So, I mentioned our algorithms getting smarter about when to ask people to pay. I mentioned our newsletters having more news today than at any other point. I mentioned how important it is to get people to download and use the app. And the registration model helps us do all those things. And we really exposed it, one, because it's now the size that we're very excited about; and two, we think it's a really important factor in the underlying model and it gives us a lot of confidence about how we grow from here.

Thank you very much.

You're welcome.

Thank you. And this concludes our question-and-answer session. I would like to turn the conference back over to Harlan Toplitzky for any closing remarks.

Thank you for joining us this morning. We look forward to talking to you again next quarter.

Thank you. The conference has now concluded. Thank you for attending today's presentation. You may now disconnect your lines.