Welcome to Spotify's third quarter 2019 financial results question and answer session. A copy of the company's shareholder letter, issued pre-market open today, is available on the investor relations website, investors.spotify.com. This call is being recorded and an archived replay will be available on the IR site after the event concludes. I will now turn the call over to Paul Vogel, Head of Investor Relations and FP&A. You may now begin your conference.
Great. Thank you. Welcome to Spotify's third quarter 2019 earnings conference call. With us today are Daniel Ek, Spotify's CEO, and Barry McCarthy, Spotify's CFO. The format of today's call will be similar to prior quarters. Daniel will give brief opening remarks followed by an online question and answer session. Questions can be submitted either through the widget alongside the webcast or by emailing directly to ir@spotify.com. We'll get to as many questions as we can. The call will last approximately 30 minutes. Before we begin, let me quickly cover the safe harbor. During this call, we will make forward-looking statements including projections or estimates about the future performance of the company. These estimates are based on current expectations and assumptions that are subject to risks and uncertainties.
Actual results could materially differ because of factors discussed on today's call and in our letter to shareholders and filings with the Securities and Exchange Commission. During this call, we refer to certain non-IFRS financial measures. Reconciliation between our IFRS and non-IFRS financial measures can be found in our letter to shareholders on the financial sections of the investor relations page of our website and also furnished today on Form 6-K. With that, I will turn it over to Daniel.
All right. Thank you. Before we discuss results, I would like to point out that today's investor letter discloses that Barry McCarthy will retire from Spotify on January 15th, 2020. During his almost five years with us, Barry played a pivotal role in our direct listing and has been instrumental in helping to establish us as a trusted public company. Barry will be passing the torch to Paul Vogel, our current head of FP&A, treasury, and investor relations. Paul is no stranger to all of you, and he and Barry have worked closely together for some time. As you may also know, before Barry stepped into his operating role, he was a member of our board of directors. Subject to shareholder approval, I look forward to having him rejoin the board in 2020.
We're pleased that Barry will remain involved with our future growth. I'm confident we will have a smooth CFO transition. I can't thank Barry enough for all that he has helped us accomplish. I'm looking forward to working with him on the board and with Paul in his new capacity as CFO. Barry's news comes on the heels of strong third-quarter results. The business met or exceeded our expectations by every measure. This quarter showed continued strong growth in users as MAU exceeded our expectations while showing another quarter of accelerating growth. Both subscribers and revenue finished slightly ahead of plan. Our gross margin finished above our guidance range. We're also pleased that we delivered positive operating income and the eighth consecutive quarter of positive free cash flow. You'll remember in February that we announced our intention to become the world's leading audio platform.
We made the strategic bet that music and podcasts are additive and that users would enjoy having podcasts as part of their Spotify experience. We're seeing evidence that this bet is paying off with signs of increased engagement and higher conversion rates from free to paid. We're encouraged, and we intend to continue to invest against this early success. Overall, we're really pleased with the quarter and the momentum we're having entering the last quarter of the year. Now, Paul, let's open the call to questions.
Great. Thanks, Daniel. Our first question comes from Eric Sheridan at UBS. On the podcast strategy, can you help us understand how assets acquired to date to accomplish the podcast strategy have been integrated with the company? How do you think about using M&A to bolster content and production capabilities compared to organic investment in 2020 and beyond?
We've really just begun the overall integration between the companies. You can see already in this quarter that we have announced a number of different originals, both from Parcast and from Gimlet. We're really pleased with the early results from those. Obviously, over time, you should expect us to increase the number of original productions that we want to make.
Next question comes from Ross Sandler at Barclays. Paul, any philosophical changes to how you plan on running things as CFO from how Barry ran things? Secondly, any additional color on the ad-supported gross margin compression in the quarter? I'll take the first part. No, Ross, obviously Barry spent a lot of time building up an organization to go public and an organization that's running really well. My plan is just keep things going as they are, and I don't expect any major changes at all in terms of how the org is run or operates. I'll turn it over to Barry for the ads question. On the ads question, the gross margin compression in the quarter, there was a couple of things year-over-year.
One would be some of the MGs around new market launches that impacted the quarter, and there was a little bit also on some of the non-music content that hit as well, which is what caused a slight decline year-over-year on the ads side on the gross margin. The next question comes from Doug Anmuth at J.P. Morgan. Can you provide us with an update on where you are with the label negotiations? What are the key outstanding factors here? Can we still expect margins on the core music business to be stable?
It's really no new updates on the timing. As we said last quarter, this is our sixth major renegotiation in our 13-year history, and there's really nothing different regarding this round. Just again, to up level what we said, we don't expect any material differences with the exception of the introduction of the marketplace strategy, which you can see that we're rolling out this quarter with sponsored listing as just one example.
I would say aside from that, we do see some scale economies in the business in the other cost of revenue line associated with reduction in streaming costs, reduction in customer service as a % of revenue, and reduction in payment fees as a % of revenue in this quarter. Those were offset by a couple of one-time non-recurring charges in the biz.
Next question comes from Ben Swinburne at Morgan Stanley. How much of the advertising revenue volatility versus expectations is due to the nature of digital audio advertising as an ecosystem, lack of standards, measurements, advertiser buy-in, or simply internal execution? Do you expect the business to match your guidance more closely in the future?
Excluding execution, none of it. That 80% of underperformance in the quarter was related to our dropping the ball on the implementation of a new operating system. We moved off of a Google ad stack that had been sunset. The core thing I would want you to know is that there was demand for that product. We were just simply unable to run it on the site. The ad business today is performing strongly. I own that miss. It's embarrassing, but it's not related to the strength of the biz.
Next question comes from Mark Mahaney at RBC. In your shareholder letter, you mentioned that a portion of the faster MAU growth result of continued product innovation. Could you speak a little more about which of those product innovations you saw work best?
Well, it's really a number of different things. One of them being we rolled out Spotify Lite in emerging markets. That is a massive thing when you're looking at data costs being a big contributor factor. We improved the quality of recommendations for new users coming to the service. We added a seven-day trial early on in the flow. There's just simply a lot of different executional things that we added this quarter, and it's that kind of pace of many small additions that then adds up to overall stronger engagement and overall then stronger retention on our sort of core user behaviors.
The cumulative effects are starting to drive faster growth in the top of the funnel. By way of example, we're seeing pretty significant increases in MAU retention, particularly if you look at it over time. We're seeing significant increases in overall engagement, and particularly if you factor in the effects of podcasts. There's lots of goodness happening in the ecosystem that's contributing to driving a virtuous cycle. Mostly to date, we've only talked about the costs associated with that, but increasingly, you'll hear us talking about the benefits.
Next question comes from Matthew Thornton at Truist Securities. Spotify has been testing a 3-month free trial, up from one month and in line with Apple Music, in between Spotify's usual biannual promotional campaigns in 2Q and 4Q. Why the test? Any lift in 3Q 2019 subscribers from the test, and what are the findings go-forward implications from the test?
Well, it didn't have any material impact on our subscriber growth in the quarter. You should really just look at this as us improving our overall customer proposition. I think this brings us more in line with what we've seen competition do as well. Most of our competition, for instance, already since the last year, were on a 90-day trial. We just brought it to where competition already is. It didn't have any material impact in the quarter.
Yeah, actually, it had no impact in the quarter, not just non-material, as compared with our original expectations. It was done for the other reasons that Daniel articulated. There will be an effect on how some of the costs of the program are reflected in the P&L as between marketing and gross margin. I'll talk to you about that offline.
Next question comes from Jessica Reif at Bank of America Merrill Lynch . Two-sided marketplace. What are the typical economics of the promotional activity that allow labels to promote new releases? Are there protections in place to protect the integrity of the user experience?
Most important thing, obviously, is the user experience and protecting that. That's why when we rolled out the product, we were also very clear with two things. One is if you're a paying customer, you have the opportunity of turning off the sponsored listing. We expect most not to do that because this is actually a widely asked-for feature by our customers. Two, in terms of how this works is, it's essentially like many other sponsored products that you would expect to see on various type of marketplaces like Amazon or Google, where there's a bidding model for it.
The next question comes from Rich Greenfield at LightShed Partners. Barry, could you reflect on your tenure at Spotify, particularly the past 18 months since hitting the public market? The stock is down slightly over that period. What is the market missing? Is there anything you would have done differently?
Well, I would've located the company in California. There are lots of parallels. There are many differences, but there are lots of parallels between the way the stock market relates to Spotify and the stock market related to Netflix. There were long periods of time when, before the stock market kind of figured out Netflix, and just like it will eventually figure out Spotify. Streaming was to Netflix as podcasting is to Spotify. There was a time when we increased the investment in streaming and Netflix at the expense of the P&L, and before we began to talk with investors about the goodness that would come as a result of the increased spending and the growth in the TAM.
In October of last year, for the first time, you heard us begin to speak about the increased investment and the erosion in margin, and we said absolutely nothing about the benefits that would be associated with the spending. Of course, the stock went down. The benefits over the long term, I think, are quite clear, and eventually the Street will figure it out. It's going to drive a virtuous cycle. The only question is how big will it be, and will it be a winner-take-all marketplace? We can see in video that streaming wins and linear dies. Same thing is going to happen in audio. That means broadcast radio is going to tip. Who eventually becomes dominant in that space is a game still to be played. At the moment, we're the largest, and I think competitively advantaged, and it's our game to lose.
The next question comes from Heath Terry at Goldman Sachs. You disaggregate the leverage you saw in core music royalty costs. Did the rise in MAU relative to premium subscribers contribute to non-music consumption? Did the new royalty agreements you signed with two of your major label partners help? Anything else?
No, actually, I'm not going to disaggregate it for you. Yes, new agreements are helping. There is leverage that we are seeing, but nothing more to say about it.
Yeah, I would say the only thing I would add, Heath, on the margin side, is just that we did benefit from premium being a little bit better than ads in the quarter. Premium gross margins are higher than the free business gross margin. When the premium business is a little bit better, it's going to help the overall gross margins, which did have an impact on the quarter.
In the quarter, that was the principal driver of the margin expansion. As it relates to the label agreements, to reiterate what Daniel has said in previous calls, and Paul and me, these negotiations are not about expansion of margin. This is about enabling the two-sided marketplace and the podcast initiatives.
Next question comes from Justin Patterson at Raymond James. How should we think about the phasing of podcast costs into the income statement? Given the exponential growth in hours and positive engagement trends, how does this change the speed at which you invest in content? Which levers are most important towards driving profitability from podcast investment?
Well, we're going to talk to you about the scale of the podcast investment next quarter. From the language we included in this quarter's investor letter, it should be your expectation that we're likely to lean into that more aggressively because of the success we're having, and you should feel good about that, one. Two, as it relates to expensing of podcast content, we first figure out what the useful life of the content is, and then we amortize the content on an accelerated basis over the useful life. Every quarter, we revisit our assumptions about useful life to make sure that they actually accurately reflect the experience we're seeing on the website. Over time, what I expect to happen is that the useful life will increase as the audience and audience engagement increases. [Tom.]
Next question comes from John Egbert at Stifel. Is it your expectation that offering 90-day free trials year-round might smooth out the seasonality in subscriber additions we've seen in recent years? Could you help us think about how to best forecast the incremental cost impact of the extra two months of free streaming? Are a meaningful portion of your gross ads eligible for three months free? It will likely smooth out the net adds across the year. We still expect there to be a holiday uptake, just generally of the buying patterns that we're seeing. I'll let Barry respond to the other question.
We're not expecting to see a significant increase in overall acquisition cost, although the buckets of acquisition costs might shift.
Next question comes from Kevin Rippey at Evercore ISI. On transition from DoubleClick, does this challenge persist until we anniversary the impact, or was this a one-time slowdown by way of transition to the new software?
Well, normally we deliver about 99.5% of what we sold. We're currently delivering 97.5%. There's some small residual tax on the business. We've built that into our forecast for Q4. My expectation is that we return the delivery rates back to 97%, or excuse me, 99 and a half.
Next question comes from Brian Russo at Credit Suisse. Regarding the price increase test for Premium Family in Scandinavia, can you update on what the response has been from your members so far in terms of both gross adds and churn? Does this move indicate any change on your thinking about price versus market growth or market share for Spotify? What other markets do you believe may have similar maturity to weather the same price testing?
Well, just let's up level again to our overall strategy. Our overall strategy is still growth. We believe streaming is a very big market, and we want a partake in that market. For the foreseeable future, most of the time spent, we're really focusing on growing. That said, there's obviously local nuances that happen that could sometimes be inflation in pricing, that could be tax-related things, and sometimes it's even down to market maturity, and in those cases, you should expect us to accordingly adjust the prices. Specifically related to the Nordic countries, it's early days. We haven't yet seen any material impact in either gross adds or churn.
Next question comes from Maria Ripps at Canaccord. On sponsored recommendations, what are some of the milestones that you would like to see before expanding the beta test to more label partners and into more countries? Recognizing that it's early, can you give us some color on how you expect the benefit to flow through your financials?
Well, it's really about artist success. The key primary notion, and you've heard me say this through prior calls as well, marketing remains one of the largest challenges that the music industry has. As there's more and more content, it's harder and harder to get your music out there and get exposed. We're really looking at that as the key success metric if we're able to break artists leveraging this tool. If we are, we think it'll be really successful, and we'll be able to roll it out more widely. Overall, the way you should think about this is, compared to our normal streaming business, this is more of a software-type margins. As such, we expect obviously, as we grow the business, that that will accrue back to Spotify in a nice way.
About the two-sided marketplace, I'd like you to think about the economic impact this way. Some of the benefits will be revenue, as Daniel pointed out, at very high margin relative to our core business. Some of them will be expense offsets in the form of, let's say, lower content costs. When we, next quarter, share with you the economic impact historically and our expectations for next year, probably we will talk to you about a gross profit impact associated with the two-sided marketplace rather than the components of revenue and expense offsets.
Next question comes from Lloyd Walmsley at Deutsche Bank. Can you elaborate on the drivers of better premium gross margins, and specifically what the product and revenue mix refers to? Is this a function of faster growth in markets with listening habits favoring lower cost content? Where is algotorial listening as a % of hours streamed today?
Sure. We're not going to get specific on the puts and takes on the gross margin. I would highlight again what I said earlier, which is in general, from a consolidated basis, we benefit when we've got more of our revenue coming from the premium side than the ads business, which helped us in the quarter. We're not going to disaggregate what the premium gross margin impact was. We haven't updated the algotorial streams in a while, but it's been reasonably consistent over a period of time.
Next question comes from Nick Delfas. Are you disappointed by the very small reduction in churn?
Actually, year-over-year was about 28 basis points. Of which, from my perspective, is a lot. I'm pretty psyched by the improvement in churn, actually.
Next question comes from Richard Kramer. With adding 30 exclusive podcasts during Q3 plus another 50 or so, is this the full extent of the monetizable podcast content on Spotify out of the 500,000 titles they quote? If creators do not use Anchor, how else can Spotify monetize third-party content? What does a premium sub get on the podcast side?
Yeah, we're really not yet focusing on monetization of podcasts. The primary thing for us has been really drive up the engagement, and as you could see in the quarter, we're now at 14% of our MAUs that's using it. We still want that number to grow, and we're still primarily focused on that given all the goodness that we think that leads to in the overall platform. We are seeing an increase in engagement and an increase in conversion to paid. We think those benefit in and of itself is very valuable, and that's kind of where the majority of focus is near term. Over time, though, obviously, we believe monetization of podcasts remains a huge opportunity, and it is something that we're going to start looking at in 2020.
Next question comes from an individual investor: Given significant elevation in regulatory focus on the largest technology companies, how do you see the potential developments creating opportunity or risk for you?
Clearly, we're moving into a different environment overall, being a tech company. As tech just overall is increasing its size and importance to society at large, so does the need for regulation. This is not something that's particularly surprising to us. It's something that we've leaned into and we've expected for some time to come. I think the good news is we're based out of Europe. We're very well aware of data protection. GDPR is something that we leaned in on from many years ago, and I feel really good about all the work that we're doing there.
I'll take another question from an individual investor. Can management elaborate the differences in driving discovery and curation for podcasts versus music? How much of your knowledge in driving music discovery could be applied to podcasts? Does the longer form nature of podcasts make it more challenging to surface relevant content for users?
There's lots of similarities and then there's also definitely a lot of things that are vastly different with podcasts. I think that the general mechanics are the same. We can certainly leverage the data that we have on our users, and we're seeing some really interesting results just in recommending podcasts based on even music listening being something that has a positive effect for us. Now, that said, selling, so to speak in quotes, music versus selling podcasts is a different proposition. One is a three-minute time commitment. Another one is an hour time commitment on average. We're still experimenting with formats and how to do that. It's still very much early days, but as you could see, quarter-over-quarter growth of 39% is incredibly strong.
We're very pleased with that and you should expect us to innovate on all forms of merchandising and discovery tools that we have so that we can drive that number up significantly.
Next question comes from Matt Janiga from Surveyor: What is the benefit of offering the Google Home Mini to existing subscribers? Will it help with churn or keep individual subs from sharing family plans?
The primary benefit really is bringing Spotify into your home. What we see is that there's three primary modalities that people engage in audio content. One is obviously on the go and mobile, where we've historically been very strong. The other one is in your home, and the third is in your car. In mobile, we're doing really well. On the car, we now have more than 70 million monthly active users who are tuning in. In the home, we look at something like the Google Home Mini plan as a great way for us to get into people's homes and get competitive against all the other services that are there. Net-net, we're expecting more engagement and more engagement typically leads to lower churn, and that's a huge benefit obviously overall to Spotify.
I'm going to take one last question from John Tinker at Gabelli: Any updates on your plans in China? What impact does Tencent investing directly in UMG have?
No real updates. We've continued to be pleased with the partnership with TME. Again, there's really no impact on Tencent's investments in Universal from our point of view.
Great. With that, Daniel, I'll turn it back over to you for some closing comments.
Yeah. In closing, we're really pleased with this quarter's results. We're really strong at every level: engagement, conversion, and retention. We will continue to improve our user experience and add value in our podcasting and marketplace offerings. We look forward to building on these successes. With that, I say thank you to everyone.
Great. Thanks, everyone. We'll speak to you next quarter.
This concludes today's conference call. Thank you for joining us and you may now disconnect.