Match Group, Inc. (NASDAQ:MTCH) Q3 2017 Earnings Conference Call November 8, 2017 8:30 AM ET
Lance Barton - SVP, IR & Corporate Development
Grégory Blatt - Chairman & CEO
Amanda Ginsberg - CEO, Match Group North America
Gary Swidler - CFO
Cory Carpenter - JPMorgan Chase & Co.
Jason Helfstein - Oppenheimer & Co.
Lloyd Walmsley - Deutsche Bank AG
Daniel Salmon - BMO Capital Markets
John Blackledge - Cowen and Company
Ross Sandler - Barclays PLC
Brent Thill - Jefferies
Peter Stabler - Wells Fargo Securities
Samuel Kemp - Piper Jaffray Companies
Mark Kelley - Citigroup
Christopher Merwin - Goldman Sachs Group
Good morning, and welcome to the Match Group Third Quarter 2017 Earnings Conference Call. [Operator Instructions]. Please note this event is being recorded.
I would now like to turn the conference over to Mr. Lance Barton, Senior Vice President of Investor Relations. Please go ahead.
Thank you, operator, and good morning, everyone. Joining me on the call today, we have Grég Blatt, Chairman and CEO; Mandy Ginsberg, CEO of North America; and Gary Swidler, our CFO. They will review the investor presentation that's available on our website and then open it up for questions.
I'd also like to remind everyone that during this call, we may discuss our outlook and future 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 earnings release and in our periodic filings with SEC.
Over to you, Grég.
Thanks, Lance. Good morning, everybody. Glad you could join us for the call this morning. We had a great quarter, and we're excited to talk to you about it. As you know, this will be my last earnings call at Match Group, as Mandy will be taking over for me as CEO. Accordingly, we thought it made sense for her to join the call now so you could hear directly from her on the North American businesses she's been running. She started getting up to speed on everything else. But obviously, that's an ongoing process. Frankly, I'm still trying to get up to speed on everything, and I've had the job for a long time.
As I've said multiple times before, working with Mandy over the last 10 or so years has been one with professional pleasures. I have the utmost confidence she'll continue the good work we'd been doing without missing a beat, and I'm sure you all come to appreciate her as I have. Frankly, from a shareholder perspective, you're all getting a hell of a trade. It's now 2 years since we took Match Group public, almost, I guess, a week to the day. Looking back, I think we've delivered on most of the things we said we'd deliver on. The company's stock performed pretty well. And as I hand off the baton, I think the company overall is in great shape. It's a great feeling, and I'm really proud of what collectively we've all accomplished.
Now to the slides. Let's start with Slide 4. Last call, I highlighted strong end of Q2 and early Q3 trends, and that continued throughout the quarter. PMC growth accelerated to 18%, driven by a great quarter for Tinder and also increased stability at Match and OkCupid. Mandy will talk about this more, but ending PMC was actually up at OkCupid in Q3, and we saw it up in October at Match.
Mandy and her team did great work at Match and Elie Seidman, Ok's CEO, really came through, delivering this turnaround ahead of schedule. Affinity declined about the same amount as last quarter, as we expected. And we think the decline should moderate from here. Again, Mandy will give you some more color on that as well.
As you've seen throughout the last 3 years, international PMC growth continues to outpace domestic due primarily to the greater mix of Tinder and the absence of Affinity.
Turning to Slide 5. Obviously, Tinder has record PMC growth in Q3, really driven by number of factors. The easy answer is Likes You goals, but it really was much more of a combination. First, we increased new users from our growth product work and marketing, both of which paid real dividends. We increased conversion, both through the introduction of the Likes You feature, but also through a bunch of optimizations we did earlier in the year, and the tech work we spent so much time on during the first part of this year has really started to pay off as well, making all this growth possible.
Hitting #1 in the App Store, I think, in August, was a big watershed for us. But the reality is, Likes You and Gold haven't really even hit yet. So there were a lot of drivers to this quarter's PMC success even those before were launched. Obviously, it's a tremendous quarter, really hitting on all cylinders at the same time.
Let's turn to Slide 6 so we can explore those revenue dynamics a little bit more. We currently have 3 ways that we monetize our users at Tinder, subscriptions, à la carte purchases and advertising. I think the right way to think about our growth modernization efforts is that we develop features for our consumers that we think will add value, but by their nature can't really be offered to all our users.
Once we've got the feature, we merchandise it in whatever way we think will maximize revenue. In other words, whether a feature is an à la carte feature or we put in Plus or we create a new SKU we call Gold; or whether we put a feature in all categories, it's really a tactical decision. Boost and Super Like, for instance, are available à la carte, as part of Plus and as part of Gold.
Take Likes You. We started with a feature. We knew that it be valuable, and then we had to decide how we wanted to merchandise it. We thought it could deliver enough value that we could charge more for it. And so we created this new Gold package, but we seriously contemplated introducing it as part of Plus. If we had, presumably we'd have more PMC than we do now because the price would be lower, but we have a lower overall ARPPU. We made the decision that trade-off was right, and we introduced Gold. Now we start testing and optimizing it over time.
The takeaway here is that we focus really on the direct revenue growth. PMC numbers, ARPPU, à la carte versus subscription pricing, particular subscription SKUs are all means to an end. And the mix of these components can and will shift over time, sometimes rapidly. We're focused on revenue, and these components are moving around all the time.
A comment on Likes You. Over the years, I think we've been pretty good at judging what features will drive monetization, but it's always hard to predict the degree of impact a given feature will have. Likes You was something we're very bullish on, and we thought we'd deliver enough value to sustain a rate increase, which is why we launched Gold. But they're just no question it meaningfully exceeded our expectations, both in its ability to drive the subscribers and to drive higher ARPPU. Really, one of the great wins we've had here.
And coming on top of the increased top of the pile dynamics that we're seeing at Tinder from the work I mentioned earlier, it's really just a great driver of financial trajectory of this business for a while to come.
Just stepping back for a second. The net effect of what we did by introducing Likes You as part of this Gold package is that we drove up conversion overall for Tinder, which means bringing in more subs. We drove up ARPPU, which obviously increases rates. But then we have a lower duration, just driven by higher pricing, which is, again, far more offset by the first two positive trends, but those are basically the directional things.
Also, just a reminder of what we call the stock and flow effect, which we've talked about before. Initially, when a new feature is exposed to the entire existing user base, you get a big surge in new subscribers. Once everyone has seen it, the surge returns back down to more normal levels, as a bigger percentage of new subscribers comes strictly from new users coming to the platform. This phenomenon has replicated itself since the beginning of time across every one of our products.
Obviously, even after that initial surge ends, PMC is at a much higher level than it was before the surge, but sequential growth, as opposed to year-over-year growth, moderates a lot. We expect this Q4 increase in sequential PMC that we will see to be similar to Q3. But then we saw expect in '18 for it to decline substantially, at least until the next major monetization push for Tinder, which we expect to come in the back half of '18.
As you know, we also drive advertising revenues through a combination of direct sales efforts in the Facebook ad network. We'll see growth this year, and even more so next year as we get the full year effect of the Facebook ads. Nonetheless, it continues to be a much smaller impact and much lower priority than our direct revenue business.
Turning to Slide 7. Despite the understandable focus on features that directly drive revenue, I know that's we talk a lot about, the biggest driver of revenue are actually the features that we don't charge for, the ones that make Tinder a desirable product to begin with, that bring our audience here and that creates a vibrant ecosystem.
The bulk of our efforts for the next few quarters will be devoted to these types of features, as the current PMC and rate surge will continue to drive significant revenue growth for at least the near term.
In particular, you're going to see a start to make changes to product that leverage what I think are a true competitive advantage in the space. These advantages really fall into three buckets, first, our users are all here to meet someone new. And this clear intent differentiates us from traditional social media products.
Second, our large scale differentiates us from other dating products. We believe there are certain product experiences that fall into something of an overlap between dating and social media, where our clear intent allows us to create an experience different from social media and our scale allows us to do it effectively where the smaller dating players could not. Playing in this area is going to be a focus of ours.
And then our third differentiator is that we just have a larger tech and product resource base than our competitors, which is necessary to hit the ambitious road map we have in front of us.
Here are just a few examples of the things we're doing, and they're by no means exhaustive. First thing we're tackling is the post-match experience. Until now, Tinder is basically a swiping machine. Once you've matched with someone, they appear on a list, and you can message with them. It's effective and simple, but ultimately limited.
In a couple of weeks, we're going to begin testing a whole new post-match experience. Effectively, it will be a rich, dynamic content experience, bringing you deep into activities of the people you've already matched with. We think it will create rich context in which to get to know your matches better, spark new and better communications, and in general, bring real texture to the product experience.
The identical experience on a social network would take on a completely different implication because of the lack of common intent, while the same experience on another dating product would be far less effective because of our higher number of matches. We're excited about this one. Of course, like all new features that might not take off, but unlike all new features, if it does, it has the potential to be transformative. And again, we'll begin testing that in a few weeks.
Similarly, in the first part of '18, we plan to begin to launch a series of location-based features, each of which builds on the prior one, over time blurring the distinction between digital and real-life dating, and dating and simply engaging in your social life. Another context, these same features could have different implications and uses, but the dating context creates a unique experience, and our scale is necessary to provide the density and depth of signal to make it effective. Again, these may or may not take off, but this vector has the potential to be defining for us.
We've talked a lot about AI before, so I won't say much. But I do want to remind that we're building a real team here, applying it to a number of areas, including in consumer-facing features that we'll be launching soon. It's going to be a meaningful part of Tinder.
Honestly, if I had to choose, the product road map at Tinder is the thing I'm most excited about at this company. The magnitude of consumer product work we're poised to engage in [indiscernible] was mostly a year of rebuilding our tech foundation to set us up to be able to launch these products which we think create a real moat between Tinder and the rest of the category.
Now I'll turn it over to Mandy to discuss North America and to Gary for financials, then open it up to Q&A.
Thanks, Grég. It's great to be on the call today and to see the terrific results across the board. Our business momentum is strong and I'm excited about taking on the reins soon. It's hard to believe that it's been 11 years since I've started at Match operating 2 brands in U.S. In fact, I joined just a few years before Grég joined in 2009. We've been in the trenches a long time together, and it's amazing at this point to step back and see the financial value we've created over the last decade as well as the impact it's had on millions of lives around the world.
Over the last few months, in addition to running the North America businesses, I've started to get up to speed on parts of the businesses that I haven't been directly involved in. My thoughts on these businesses and the portfolio overall will continue to develop, and I'm sure I'll talk about that on the next call. Suffice it to say that I'm currently excited about the opportunity in front of us.
The trend in the North America businesses have been a big focus since our IPO, and I'd like to provide my perspective. I'll talk about financials first because it's a business that has continued to see growth over this period, and I'll turn to the others reflected on Slide 8.
PlentyOfFish is led by Hesam Hosseini, a Match veteran, who took over the role about 2 years ago. It's clear that PlentyOfFish has carved out a unique space in the market and is able to generate more conversations than any other dating app. Their user base is geographically dispersed throughout the U.S., which is complementary to our other products, for example, Tinder and OkCupid tend to be more concentrated in bigger cities. PlentyOfFish has made real progress in modernizing the mobile experience this year, recently relaunching our app and adding features like Spark, which allows users to strike up a conversation straight from a profile. These efforts around product improvement and differentiation are focused on getting people into more and better-quality conversations. This initiative will continue in 2018. This has been a great acquisition for us, and Hesam has done a fantastic job.
Turning to Slide 8. I want to start with Match and OkCupid. They have been on a multiyear journey to remake themselves into mobile-centric businesses in order to reverse PMC declines. Our first step in this journey was to fix conversions by making sure that these products worked on mobile as well as they did on desktop.
Part of that mission and our primary focus for this year was to align the user experience for each of these products with their core brand proposition. We've set out to create differentiated product experience for each brand as a way to drive higher engagement that will ultimately translate into increased top-of-the-funnel activity. I'm pleased to say that we have now stabilized PMC at these 2 businesses.
Start with Match. The Match PMC momentum we saw in June and July continued throughout Q3, and we're able to grow ending PMC on a year-over-year basis in October for the first time since August 2014. I just want to stop and pause on that for a second. That's obviously something we feel great about.
I expect average PMC in Q4 to be up slightly year-over-year as we've previously anticipated. Our continued focus on products paid off, and we're still driving conversions higher, and we remain focused on rolling out features that provide our users with better ways to make real connections.
In August, we launched Match Stories, where users can create video profiles, more than 50% of our active users on apps have these profiles. We continue to develop these features with a focus on giving singles new and more reasons to try Match. I'm proud of the work the team has done to stabilize the business with these product changes.
Our next step for Match is to translate products, performance improvements into growing the top of the funnel. TV advertising has traditionally been a great way for us to reach our core audience with our message, but the media landscape is shifting quickly and TV viewership is down, making TV a less efficient channel for us.
That said, we are making progress shifting our customer acquisition efforts to where users are spending their time and reaching them through online videos through deeper storytelling. The ability for us to drive growth at Match and to what extent really depends our ability to effectively craft this practice marketing transition.
Let's turn to OkCupid. Their product work is similarly led big improvements in conversion and retention, especially with females, where retention after the first week is up 25%. This year, OkCupid redesigned their mobile experience, revamped the profile, adding back the iconic questions that had previously been buried in the mobile experience. As a result of these improvements, OkCupid achieved year-over-year growth and ending PMC in Q3, which was ahead of our schedule. This will lead to average PMC up slightly in Q4. Again, this is worth a pause.
On the last earnings call, Grég mentioned that OkCupid was a bit behind the other brands, but they caught up this quarter. Elie Seidman and the team at OkCupid deserve huge kudos for the excellent progress they have in this business, which still resonates well with its core demographic and now has a much better product to serve them.
With significant improvements at OkCupid, the attention really continues to shift to growing the top of the funnel. In its early days, OkCupid was [indiscernible] to the PR buzz from telling compelling data stories as well as taking a stand on issues. The OkCupid team is putting the brand back in that spotlight up after a few years of being quite by upping the PR game. And you can see in the slide, we are getting significant media pickup across a number of TV and online media outlets from Vice to The Tonight Show. Because OkCupid isn't backed my meaningful marketing spend, strong product experiences and positive press mentions are critical to grow the top of the funnel, and we made great strides in this business.
Lastly, I wanted to touch a little bit on the Affinity brands. As background, Affinity consists of 29 demographically focused brands operated on a common platform. The vast majority of revenue PMC comes from two of those brands, OurTime, which targets users over the age of 50, and BlackPeopleMeet. 21 of the brands focused on these demos and have very small number of PMC.
Previously, we've explained that we reduced marketing spend at Affinity because we discovered we had overstated LTV, and as a result, overspent in prior periods. The LTV declines were driven primarily by mobile shift, which hurt advertising revenue per subscriber, which was an especially big driver in this business. Lower marketing spend in turn drive lower PMC, which is what you're seeing now.
We believe we have stabilized our marketing spend to the appropriate levels to where this business is today, but we expect to spend cuts we made this year to continue to be a drag on PMC going forward, although the pace has started to moderate and should continue to do so over the next quarters. And just to take a step back. This is a small contributor to the North America businesses, and obviously, a very small part of the company overall.
I'm glad to be on the call to talk about the businesses that I'm so proud of and soon to be taking on the helm of this incredible company. We got a lot of momentum, and I'm confident that we're going to hit the ground running in 2018.
And with that, I will turn the call over to Gary to take you through financial performance.
Thanks, Mandy. Slide 10 has our key financial results for the quarter. Total year-over-year revenue growth was 19%. We were expecting revenue growth to accelerate as the year progressed following a major product push at Tinder, and that is coming to fruition. Direct revenue grew 21%, driven by 18% PMC growth and ARPPU that was up 1%. Our international business grew direct revenue 40%, driven by 33% PMC growth; while our domestic business grew 10% in direct revenue and 9% in PMC. International now accounts for 43% of total direct revenue.
We had $0.5 million decline year-over-year in indirect revenue as we continue to experience lower ad revenue at our non-Tinder brands. This was largely offset by the continued ramp of the Facebook ad program at Tinder, which delivered as we expected. As Grég discussed, Tinder contributed very strongly to our results for the quarter as PMC growth was exceptional, à la carte continued to be very strong and the introduction of Gold led to higher ARPPU at Tinder.
EBITDA grew 12% in the quarter due to the revenue growth and lower sales and marketing expenses as a percent of revenue as the mix of our businesses continue to shift to lower marketing spend brands, offset by higher Tinder operating costs and higher IAP fees. We also incurred $11 million of payroll taxes related to the former Tinder option exercises and some associate professional fees. Excluding this $11 million of cost, EBITDA growth would've been 22%, 3 points better than our 19% reported total revenue growth. Operating income was flat this quarter primarily for two reasons, we had an increase of $9 million in noncash comp, which is primarily related to Tinder and roughly half of which will not recur; and we had a $5 million reduction in continued consideration income. These were partially offset by the lower amortization of intangibles from our PlentyOfFish acquisition as a scheduled amortization from that acquisition concluded at the end of 2016.
Operating costs and expenses were 73% of revenue compared to 68% in Q3 '16. This increase was driven by the payroll taxes, professional fees and the increase in noncash comp that I mentioned previously. Sales and marketing expense was 28% of revenue versus 31% in Q3 '16, despite increases in marketing spend at Tinder, as the mix of our businesses continue to shift the lower marketing spend brands and we reduced marketing spend at our Affinity business.
Slide 11 shows our ARPPU, which overall trended up in the quarter. Total ARPPU was $0.54, up from $0.53 in Q3 2016. North American ARPPU was down only about $0.2 for the quarter as Tinder PMC comprised a large of our normal American business and we saw a shift to longer-term packages in some of our non-Tinder businesses in North America, which did impact rate slightly. This was all more than offset by a meaningful rate increase at Tinder North America, which was driven by continued strong à la carte and the initial impact of Tinder Gold.
International ARPPU is up $0.02 overall and $0.01 on a constant-currency basis, driven by the same Tinder dynamics as in North America and strength in our Japanese and European businesses. Overall Tinder ARPPU increased by approximately 25% year-over-year, driven by increased à la carte and the initial effect of Gold.
Turning to the next slide. As we've discussed, we converted Tinder options to Match Group options in July. Following the conversion, a larger number of these option holders sought liquidity, which wasn't surprising since they've held illiquid Tinder options for a while. When the exercised, we withheld shares to cover the exercise price of the options and employee withholding taxes, which we paid from cash on hand. We also had the opportunity to purchase a portion of this vested stock awards for cash.
In aggregate, we spent $500 million to buy back the vested awards and a portion of the vested awards and pay all employee taxes on the exercises, which avoided issuing a total of 26.7 million Match Group shares. The effective buyback price averaged $18.86 per Match share, which compares very favorably to last night's close of just under $27 a share. The cash mostly came from cash we had on hand, but we also increased the borrowings under our term loan by $75 million. We issued 10.6 million shares to settle options that were exercised by employees that we did not purchase for cash.
After all of this, we had $158 million of cash on hand at September 30. Our gross leverage ticked up slightly to 2.9x, and our net leverage went from 1.6 to 2.5, all still very manageable. We expect to be able to delever from these levels once again, given that our business generates strong cash flow. In fact, for the 9 months ended September 30, net cash flow from operations increased 17% to $230 million, and free cash flow increased 31% to $208 million. Free cash flow conversion from EBITDA through Q3 was 66%.
After the quarter ended, we sold our minority stake in Zhenai, a Chinese dating company. That yielded us about $60 million of cash and a $9 million gain. Overall, our financial flexibility is very strong.
Given the strong debt markets, we've been able to take actions over the past 12 months that have driven down our interest cost by approximately $7 million annually. Most recently, we lowered the rate on our term loan by 75 basis points, which yield $3 million of annual savings. We believe there may be additional actions we can take to continue to further reduce our debt costs.
Turning to Slide 13. We had a $260 million corporate income tax deduction in the quarter related to option exercises, most stemming from exercises following the conversion of Tinder options to Match options. This led to a $226 million income tax benefit in the quarter. As a result, our net income attributable to shareholders rose 410% to $288 million, and our diluted EPS jumped 367% to $0.98. Much of this relates to how employee-exercised stock-based awards are treated under the new accounting pronouncement.
In prior quarters this year, we've seen increases in diluted share count, which had driven down our diluted EPS. This quarter, we saw the tax benefit flow through net income as required under ASU 2016-'09, which in turn drove up our net income. We now have a $246 million deferred tax asset on the balance sheet. We don't expect to be a significant U.S. cash taxpayer until 2020.
Slide 14 lays out our financial outlook. Overall, we had strong momentum from Tinder and our other businesses are contributing as expected. We expect Q4 revenue of $355 million to $365 million or 22% year-over-year growth at the midpoint, a further acceleration from the 19% Q3 2017 growth rate. We also expect Q4 revenue to be positively impacted by the full quarter impact of the higher Tinder Gold rate, growth in Tinder average PMC and continued growth in our ala carte.
We expect $147 million to $152 million of EBITDA in Q4 or approximately 17% year-over-year growth at the midpoint. We are experiencing strong price momentum at Tinder and our other businesses, and we expect to spend up on marketing if we can find the opportunities. We are currently projecting a 20-plus percent increase in marketing spend year-over-year in Q4, much of it at Tinder, but also at many of our other brands.
It tends to be a little tougher to spend marketing dollars in Q4. If we find opportunities to spend, we'll do so, and that will drive EBITDA to the lower end of the range. If we find fewer spend opportunities that made sense, EBITDA will come in toward the higher end of the range. Q4 EBITDA margins would be up 42% at the midpoint of our current outlook.
Tinder's sequential Q3 increase in average PMC was above our expectations. About half of the increase stems from the stock effect of Tinder Gold, which Grég talked about, whereby current users become paying subscribers due to the new feature. This tends to happen in a period immediately after introduction of the new feature, then the impact dissipates fairly quickly. Because we rolled out Gold late in Q3, we saw a burst of PMC late in the quarter. The timing of this surge impact average PMC for Q3 and will also benefit Q4 average PMC.
As a result, we expect Q4 to be similarly strong in terms of the sequential increase in average PMC. After Q4, the stock effect will dissipate, and sequential average PMC increases will slow. We also expect Match Group PMC in Q4 to benefit from year-over-year stability at Match and OkCupid, which we are finally seeing.
We now expect revenue above our prior expectations at $1.307 billion to $1.317 billion for full year 2017, representing 17% to 18% growth over 2016, driven primarily by Tinder's strength. We expect EBITDA to be $463 million to $468 million for the full year, 15% to 16% growth over to 2016, and margins of about 35% at the midpoint. Excluding the costs related to the Tinder option conversion exercises, which have totaled $15 million for the full year, year-over-year EBITDA growth would be 19% to 20% and margins a little over 36%.
We are currently in the process of finalizing our strategic plans for 2018, so I don't want to comment too much yet on where we think next year is going to come out. But I do want to highlight some key factors to consider when looking out into 2018.
We expect year-over-year Tinder PMC growth to remain very strong through next year. The end of the Gold stock effect, however, will lead to lower early 2018 sequential average PMC increases than we have typically seen at Tinder. After that, sequential average PMC increases should return to more typical Tinder levels.
We expect to deliver a very strong top line growth again in 2018 at Match Group. We believe we are positioned for mid-teens revenue growth next year as Tinder's strength continues and the other brands are stable. Overall, growth won't quite be at the exceptional exit rates we are seeing in late 2017, but Tinder's product work this year should set us up very well for 2018, providing revenue momentum early in the year, which we expect to supplement with additional monetization features later in the year. As with the case this year, new product features that will be unveiled in 2018 will set us up well for 2019.
Tinder also has meaningful operating leverage as we scale the business. We do expect to incur some increased regulatory and other operating costs across the company in 2018, including related to the new data privacy regulations in Europe. Even with these, we expect 2018 Match Group margins to be above 2017 levels. We'll provide more details on all of this on our next earnings call.
We will now answer any questions you may have. Operator, please open the line to questions.
[Operator Instructions]. Our first question comes from Douglas Anmuth of JPMorgan.
This Cory Carpenter on for Doug Anmuth. Two questions on Tinder, maybe starting with Gold. Could you talk about the mix of adoption you're seeing between existing paying subs and also the conversion of nonpaying subs? And then any metrics you're able to share in terms of Gold subs as a percent of overall subs or early views on what you're seeing in terms of renewal rates? And then secondly, could you give us an update on the usage of Tinder's à la carte features in the quarter and any interesting trends you saw, such as may be Gold impacting the usage of Boost and Super Likes?
Thank you. Look, we don't give out a lot of that granularity. I will say that the way we think about it is the introduction of the Likes You feature itself increased conversion, drives up subs. The introduction of the Gold package drives up ARPPU, but makes that increase in conversion less than it would have been otherwise. You have 2 components going. You have both new people coming in and taking Gold. You also have an upgrade component. On virtually every one of these metrics, adoption exceeds our expectation. So mix is higher, conversion is higher, et cetera. But you really see that reflected in, I think, Gary talked about around a 25% in Tinder ARPPU, and you can see that -- Gary said that half of the -- this upsurge came from this, so you can sort of back into we don't give conversion and mix numbers.
Renewal rates were down on Gold versus Plus, which you would expect whenever you make a pricing decision. It was well within the sort of constraints that you would expect on that sort of a pricing increase. As I've said in my remarks, we will iterate on these things over time. You set out one premium, those premiums will be tested and iterated on and all the rest. So these things will move around a little bit, but that's all baked into, obviously, the outlook that Gary walked through. In terms of à la carte, I think we've talked about the fact that à la carte tends to come mostly from people who are also PMC. Therefore, when you see a surge in PMC, you tend to see a surge in à la carte. I think that has played itself out this quarter. People who use Gold will use Super Likes and Boost a little differently, so there is a toggle in terms of how they use those things. But overall, not a meaningful mix shift in those and relatively steady.
Our next question comes from Jason Helfstein of Oppenheimer.
Two questions. So with Tinder, is it fair to say this is the first time you have multiple kind of pricing tiers basically on a product? And as you're going through this -- and clearly is learning because this is new and the uptake has been clearly better than expected. Are there learnings in Tinder and possibly apply that to other products? And then just thinking about kind of how that affects conversion rates, where, historically, we've always thought about, on Tinder, the paid conversion rate would have to be relatively low, but with different pricing tiers, can you have more people on a basic price and then you have this premium price? And then second question, just on international ARPPU, can you tell us what the change was ex currency quarter-to-quarter?
Yes, Jason. Actually, it's sort of reversed. Meaning, we took all the learnings from the other businesses and applied it to Tinder. We've talked for a long time about the fact that we have effectively a dynamic pricing program. Across our other businesses, we have add-ons, we have packages, we have discounting and premium pricing and à la carte. So we've actually done this far more intricately in our other businesses than we have at Tinder. This is really the beginning of doing it at Tinder. In fact, even before Gold, we do have multiple price points at Plus, which we've talked about. They're geographically-driven, they are driven by a number of components, there's discounting. So our ARPPU numbers always include a whole variety of pricing tiers. I do think it will continue to develop at Tinder. We're far earlier along that trajectory at Tinder than we are at -- certainly, businesses like Match or Meetic, where we're doing this for a long, long time.
And so we always think about conversion both on a steady-price basis and on a price-change basis. So yes, over time, what you try and do is you have operatives at a variety of price points, you're trying to get each cohort to pay as much as they will pay. And you have to do that sort of very carefully, right? Because otherwise, if you have a simple one-price subscription package, you're almost, by definition, undercharging your entire subscription base and you are leaving behind tons of people who would pay you something. And our core merchandising program is designed around creating multiple tiers to do that. Sometimes, they had individual SKUs like Gold versus Plus. Sometimes it's just dynamic pricing where you're charging people in this place X and you're charging people in this place Y. And you're charging this person who's been here for 10 days without converting yet, and you start to discount. So there's all sorts of things that go into it. It's very intricate.
I'd like to think it's been done very sophisticated -- in a very sophisticated way than some of our other businesses. It is still very much a blunt instrument at Tinder, and one that I think that has a lot of runway for us. I think I've covered all except for your last question.
The question on the ARPPU, Jason, which is, is if you look at Slide 17 towards the back of the deck and it's also in our earnings release, you have the detail on ARPPU. But the short answer is international ARPPU is up from $0.50 to $0.52, so $0.02 increase, $0.01 was attributable to FX and the other half -- $0.01 was attributable to improved rates. So that's the breakdown there, again driven by Meetic and our Japanese business primarily.
Our next question comes from Lloyd Walmsley of Deutsche Bank.
Wondering if you guys can elaborate a bit on how some of the non-modernization product innovation drives engagement and revenue over the longer term? And how you kind of frame up the benefits to that more broadly? And then more specifically, as you look to do more post-match features along those lines on Tinder, do you see that kind of expanding the brand proposition and drawing in new demographics? Or is it more you think like they just drives deeper engagement from core demos? Any thoughts you could share there would be helpful.
Sure. We think about the business in sort of two ways, there's the product experience that we offer people for free, right, and that experience is what creates our user base. People tend not to come for the paid features. They come for the base experience, and then they pay for the sort of extra benefits on it. So at the end of the day, we need to have a great product that is both bringing new people into the category and taking share or keeping share from the rest of the category. Tinder's done that very well. But we see -- one of the reasons we did a lot of the work we did this year on the technology platforms and everything else is the ability to expand that.
And by creating differentiated products that both are more effective at delivering, which can be measured in a variety of ways, and more enjoyable to use, et cetera, all are what drives retention, word of mouth. They are the things that we market and will continue to market going forward. So we think of that as sort of the heart and soul of what we do. Once we're doing that successfully, we're bringing more and more people into the ecosystem, we are keeping them longer and that allows us to get more and more people to pay for these features. If you assume a fixed conversion rate, which obviously we don't have, the more people you bring in, to more revenue you make. So at the end of the day, this is what drives the top of the funnel at a company like Tinder.
In terms of post-match, post-match is really, I think mostly about creating a richer, more enjoyable, more engaging experience that's more effective that I think will drive greater customer satisfaction, greater word of mouth, et cetera. I think some a stuff we're thinking about in location and some others areas are sort of really outwardly sexy and have the potential to bring in sort of new audience even beyond the categories. So I think we're trying to play both in sort of the heart and soul of dating and also sort of continue to expand definition of it with Tinder. This past year was a lot of blocking and tackling on the tech side. We then did launch some new features.
We launched a modernization of our navigation, which a lot of people move through profiles and photos much easier. We launched something called Reactions, which is sort of a fun language for sort of communicating back and forth. And those are all sort of things that we've launched that we think are fun and good. But I think some of the things that we're more moving towards right now are much more sort of potentially transformative and really, I think, can really move user growth meaningfully.
Our next question comes from Dan Salmon of BMO Capital Markets.
So maybe Grég and Mandy, just take a step back with the launch of Gold here. Do you, at a high level, take maybe a different view on what your long-term potential is for PMCs as a percentage of your monthly active users as it's even more traction that you think long term that you can convert to paying users as you add more value and add different tiers? That's the first one. And then maybe just a quick update on the Tinder leadership transition.
Sure. On the Tinder part of it, look, Tinder has, frankly, exceeded our expectations sort of at every step. So in terms of rethinking our -- I wouldn't say we've -- I mean, there's obviously some ceiling of the percentage of payers, of users that will get to pay. We don't really have a hard view on what that is. We keep rolling out new things, and they keep buying them. And I think we've done a good job. And I think there's still a fair amount of runway there to go, but we certainly don't put a number on it. Mandy, you've obviously seen the other businesses, they're somewhat developed, but you've been continuing to drive conversion as well. So...
Yes, for a lot of other businesses we've worked on, as you've heard, it's been a big focus on product innovation. And we've seen real stability in PMC as a result of it. So for us, it's -- we again don't see a ceiling, although it's certainly more and more challenging as time progresses for the more traditional businesses to see that lift.
In terms of Tinder leadership, obviously, there's been a process. It's the kind of thing where until there's an announcement, there's not really an announcement. But we feel very good about where we are in bringing that to fruition.
Our next question comes from John Blackledge of Cowan.
Great. On Tinder advertising, you mentioned on the deck that it increased. Just wondering how the partnership with Facebook is going and how we should think about the advertising revenue trajectory into 2018. Do we think about it as kind -- as perhaps a step function uptick? And then secondly, on Tinder. I think you ramped marketing expenses in some emerging markets, like India, to drive adoption and viral user growth. Just wondering kind of how does marketing efforts are going. And for the Tinder marketing spend in 4Q, is that spend intended for those emerging markets? Or is it more broad based?
Okay. The partnership with Facebook is going great. Obviously, I think we have less than -- '17 will reflect the less than even half year sort of run rate. Next year we'll get the full benefit of that, and so we will have ad revenue increase next year. I'll just repeat what I've said before, which is in light of just how strong the direct modernization is, we are not -- we are doing that and we are growing that revenue, but it is not a major focus of ours. We are not -- on a relative basis, we're not sort of pushing the envelope in terms of ad frequency or that sort of thing. And so given that, sort of once the network is -- once we've anniversaried that, you sort of will grow it modestly, unless we choose to get more aggressive on frequency, and that's certainly not a decision that we're making, I think, in the near term. In terms of marketing efforts, yes, we have -- we stepped up marketing a lot this year much of it international, some of the domestic.
I think it's been very effective. I still don't think we've even hit our stride yet in doing it. I think we certainly spent far less than other big players in the space. We've got great growth that's coming both through monetization and through user growth. We expect that marketing to continue and increase again next year. But obviously, as a percentage of overall revenue, which we expect to grow very much next year, it's not growing and, again, I think remains far smaller than, really, the competition.
Our next question comes from Ross Sandler of Barclays.
Great. Just one for Mandy and then one for Gary. Mandy, just, I guess, stepping back in philosophical changes now that you're taking the baton from Grég around growth margins or the company's use of capital or should we just expect more of the same under your leadership? And then Gary, on the PMC cadence. So the 476,000, if you kind of look at how you guys report and the fact that that's an average, it was very back end-weighted. So the fact that you're guiding the same number in 4Q, just curious what does the October trajectory look like versus September? And why wouldn't 4Q be way higher than 476,000 given that you only had 2 weeks on Android and 4 weeks on iOS in the third quarter?
I think on the PMC side, I think it's important to understand -- and Grég alluded to some of this, we have the big effect of Gold and you're seeing that and that came late in the quarter. But we also have a lot of momentum leading into the quarter, so that's all reflected in the Q3 numbers. And so even in late Q2, you had seen stronger PMC growth than what you have seen kind prior to that. And so the way that, that kind of math all works out, ultimately, when you take -- look at average PMC is you get this big burst of average PMC increase in Q3, and we are expecting something that will roughly replicate that again in Q4. That's going to be kind of how the math works out. So it's not just that Match has surge from Gold. Obviously, that contributed especially in Q3, as you're saying, but it will start to peter out towards the middle of Q4. And so the surge effect is visible in average PMC increase in Q3 and Q4, and then peters out again after the end of the year.
Our next question comes from Brent Thill of Jefferies.
Just on Tinder Gold, curious if you could give us a sense of what you've seen in terms of monthly versus 6 months versus annual sign-up out of the gate.
I don't think we really give that information in terms of package mix. And I think it's still developing at this time. So -- yes, sorry, I don't think we really give that. I think that, over time, we expect it to be -- it's not nearly that we expect them to get meaningfully from Plus, but I don't think we've fully sort of merchandised all the packages in that way yet.
Okay. And just for Mandy, just -- when you think about -- you mention building on the base. If you had kind of a single strategic priority as you look that you'd like to achieve going into '18, what -- how would you articulate that?
Gosh, I mean, I officially haven't taken over the job yet, and I'll -- as I get up to speed, I'll develop more sort of my strategic intent. But I think for me, who's been in the business for a really long time, it's -- you've seen tremendous growth at Tinder. It's going to be our growth engine and we'll continue to invest in that growth. And then really around as we stabilize PMC in a lot of the other business, it's building on top of that and cracking the top of the funnel.
Our next question comes from Peter Stabler of Wells Fargo.
A high-level question for Mandy. Now that you've completed a lot of the heavy-lifting on refashioning the non-Tinder assets for mobile, wondering if you could talk a little bit, philosophically, about hard versus soft paywalls, and whether you think there are any significant market shifts that of occurred which could, I guess, challenge the hard paywall model going forward. Or is it really just a question of feature sets and making sure that you're addressing the individual consumer needs with those products?
So I think, especially if you look across the U.S. products, which I know well, I believe in the portfolio strategy. I mean, each product offers a very different value proposition. A 45-year-old single mom in Austin is going to be very different than a 25-year-old person who just moved to New York City. And so -- and people will pay for value. When it comes to Match, for example, it's really around the quality and intent. So I think, for us, just having this multiproduct portfolio strategy makes sense. And the business will evolve, but we feel pretty good about this mix of both the hard paywall and soft paywall business.
Our next question comes from Sam Kemp of Piper Jaffray.
Mandy or Grég, you guys launched Tinder browser payments during October. Can you talk about whether or not you've tested funneling people away from the app to avoid the App Store tools, and maybe just broader thoughts on the opportunity to do that? And then Gary, just one on incremental margins. When we look at Tinder, it's obviously got something probably close to 75% gross margin given the App Store fees. But when you think about the flow-through of new gross profit down to EBITDA, can you just qualitatively talk about what impacts those dynamics and whether or not that's been improving or stabilizing?
On the webpages for Tinder, obviously it, as in other businesses, we've had the experience of having both web payments and app payments for a while and the developing strategy there. Obviously, you want to optimize that as much as you can. That 30% is a heavy tax. And so over time, we developed the strategy. They are, as you might imagine, somewhat competitive. And certainly, at Tinder, where we're just sort of rolling that out, we haven't really engaged in that in any meaningful way yet. But certainly, it is a tool that we'd like to use over time to try and help our margin.
When you think about Tinder margins, you're right, 70% gross margins, I mean, expenses being marketing and headcount after that. And we're refining our thoughts on '18 at the moment around that. But clearly, there's natural leverage there. We've hired a lot of people over the last little while. Obviously, these developments -- these developers are expensive. And so we're trying to figure out in a competitive market how to handle all that. And then you got the marketing spend. We ramped up marketing spend very significantly in '17. It's been pretty successful as Grég talked about.
We are trying to figure out the right amount to kind of reinvest from that marketing spend into '18. But we're going to continue to try to push that because we think, long term, that's beneficial for Tinder. So we think we can still show good margin improvement at Tinder and overall and still put out the requisite amount of spend that we need for that business. And that is our plan. So we're still calibrating that. We'll give you a little bit more detail on it next call. But the trends are the ones that I spoke about, which is we think there's leverage at Tinder. And we think that'll drive operating margins overall for the company.
If I can, a quick follow-up on that. Can you talk about the split of marketing spend at Tinder for core geographies versus kind of new footprint expansion?
I would say it is -- it's roughly proportional to users, a little bit heavier on the user base in developing markets, but it's roughly proportional.
Our next question comes from Mark Kelley of Citi.
First one is the comments on PMC expectations for Tinder next year, super helpful. I'm just curious, is there anything you can point to historically on the ARPPU side after the surge? Has that stayed kind of similar run rates? Or does that come off at all? And second, given of all the international commentary you've made on Tinder, should we assume that international Tinder PMCs were maybe a little bit more of the mix of this year and maybe that the PMC growth outpaced that 25% you've commented on?
On the Tinder PMC and the ARPPU, our expectation is that you're going to continue to see relatively strong growth in Tinder ARPPU as the stock effect holds. We think we'll be able to maintain close to that, but you might see a little pressure as that effect dissipates. Obviously, that's kind of what we do product wise next year, so it's a little bit early to comment, but those general trends are kind of what we're expecting from an ARPPU perspective at Tinder. And then a question on the international PMC, could you just run that past us one more time?
Yes, sure. I know the first test of Gold was really outside the U.S. or outside of North America. So given that, given the increased marketing spend internationally, should we assume that maybe the mix of Tinder PMC skews a little bit more to the international side this quarter? And that maybe the ARPPU growth of 25% you talked about, did international outpace that?
I don't think there's been a meaningful shift in the geographical mix of PMC or ARPPU. I mean, I'm sure it varies modestly, but not in the way that we have focused on. And then just going back to the earlier answer that Gary gave, I just want to harken back to an answer I think I gave to Jason's question earlier, which is we are increasingly, as Tinder gets more sophisticated, going to be toggling between rates -- between subscription rates, package mix, à la carte, different SKUs, trade-offs between conversion and pricing, which we've done so much in our other businesses.
So I think a lot goes into ARPPU over the course of the year. And it's just -- for better for the business, but frankly harder, I guess, from a modeling perspective. It's very hard to just take a thing and sort of run rate it over a long period of time. So that will move around a lot. And we don't even know necessarily today exactly which trade-offs we'll make because we'll be testing them throughout the year. So I think when Gary gives that sort of top line revenue guidance, in that is a whole bunch of variations that play between these various metrics.
Our final question comes from Chris Merwin from Goldman Sachs.
So just for the core business, you called out PMC being higher year-on-year in October, I think for the first time since August 2014. You also talked about improving trends for the core in the fourth quarter. How much of that is a just underlying momentum in the business as compared to a step-function increase from product updates, like video when that's rolled out? And then just secondly of marketing for next year, you also talked about a change in the strategy on Match as you move dollars from TV to online video. Are you assuming the same efficiency in your marketing going forward? Or could there be any improvement there that we could expect to show up in 2018?
Okay. So regarding PMC, a lot of the gains that we've made in PMC really have been a function of improved product experience. And when you improve product experience, you have higher engagement and so it helps the ecosystem. So it's hard to sort of isolates organic versus what we do because it's all impacted. The second question that you asked was around marketing. I think TV advertising has been an efficient channel for us, and we have to go where the eyeballs are. The good thing is that a lot of the -- there's a lot more video ad products that we're leveraging to be able to tell that longer-brand story. But there's still a delta between TV and online. I mean, TV generally tends to be a more efficient channel in the past.
All right, guys. Thank you all very much. My last call. To all the shareholders out there for whom I work, it's been a great pleasure. And I have great confidence that Mandy is going to kill it going forward. And in my forthcoming role as Vice Chairman, I will do everything we can to help make sure that, that happens. So very confident. Feeling great about where the company is. And thank you all for your support.
The conference has now concluded. Thank you for attending today's presentation. You may now disconnect your lines.