Pandora Media (P) Timothy Westergren on Q2 2016 Results - Earnings Call Transcript

| About: Pandora Media (P)

Pandora Media, Inc. (NYSE:P)

Q2 2016 Earnings Call

July 21, 2016 5:00 pm ET

Executives

Dominic Paschel - VP, Corporate Finance & Investor Relations

Timothy Westergren - Chief Executive Officer & Director

Sara Clemens - Chief Operating Officer

Michael S. Herring - President & Chief Financial Officer

Analysts

Heath Terry - Goldman Sachs & Co.

Benjamin Daniel Swinburne - Morgan Stanley & Co. LLC

Stan X. Meyers - Piper Jaffray & Co. (Broker)

Douglas T. Anmuth - JPMorgan Securities LLC

Mark Mahaney - RBC Capital Markets LLC

Jason Helfstein - Oppenheimer & Co., Inc. (Broker)

Sameet Sinha - B. Riley & Co. LLC

John Egbert - Stifel, Nicolaus & Co., Inc.

Nat H. Schindler - Bank of America - Merrill Lynch

Matthew C. Thornton - SunTrust Robinson Humphrey, Inc.

Mark Kelley - Citigroup Global Markets, Inc. (Broker)

Michael Graham - Canaccord Genuity, Inc.

Barton Crockett - FBR Capital Markets & Co.

Ronald V. Josey - JMP Securities LLC

Rich R. Tullo - Albert Fried & Co. LLC

Peter C. Stabler - Wells Fargo Securities LLC

Operator

Welcome to Pandora's Second Quarter 2016 Financial Results Conference Call. All lines have been placed on mute. There will be a question-and-answer session at the end of the conference.

Opening today's call is Dominic Paschel, Vice President, Pandora.

Dominic Paschel - VP, Corporate Finance & Investor Relations

Thanks, Mike. Good afternoon, and welcome to Pandora's second quarter 2016 financial results call. Before we begin, let me remind everybody that today's discussion will contain forward‐looking statements based on our current assumptions, expectations and beliefs, including projected financial results or operating metrics, business strategies, anticipated future products or services, anticipated market demand or opportunities and other forward‐looking topics. For a discussion of the specific risk factors that could cause our actual results to differ materially from today's discussion, please refer to the documents we file with the Securities and Exchange Commission.

Also, during this call, we will discuss non‐GAAP measures of our performance. GAAP financial reconciliations and supplemental financial information are provided in the press release filed today with the SEC, and detailed financials are available on our Investor Relations site. Today's call is available via webcast and a replay will be available for two weeks. We will also post the full text of today's prepared remarks once Mike concludes. You can find all of the information I have just described on the IR section of Pandora.com. On today's call, we have Tim Westergren, Founder and CEO, Mike Herring, President and CFO, and Sara Clemens, COO. With that, let me turn the call over to Tim Westergren, Pandora's CEO and Founder.

Timothy Westergren - Chief Executive Officer & Director

Thanks, Dom, and thank you for everyone for joining the call today. We continue to make great progress toward our vision of building a complete music marketplace that satisfies the full range of music listening for the consumer, and provides a complete solution for artists. The product is really coming together. As we look to build the next generation Pandora music service, it's clear that just as we did with the launch of personalized radio, we will be bringing a unique product that will reinvent the category. The size of our audience and the immense amount of preference data we have for the approximately 100 million listeners that visit Pandora every three months will enable us to approach the listener experience in a completely different way: intuitive and easy to use.

In Q2, we continued to execute on our plans, while crossing several important milestones. We continued to have very productive negotiations with the three major labels, as well as with our partners in the independent label community. Although these discussions are confidential, we believe they will result in our being able to deliver the product we want with the right business model, and that the investments we are making today will lead to win‐win results as they are delivered, both for Pandora and the music industry as a whole.

Top‐line revenue was slightly below expectations in Q2 due to softness in national advertising in the entertainment and telco sectors. Although hours reaccelerated, there is a delay between when the inventory is realized and when we can monetize it. We are confident in our ability to more effectively capture this incremental revenue over time, as we continue to see strong performance and demand for our advertising products.

We feel more confident than ever about the core advertising business due to: one, strong local advertising results, now representing 28% of our revenue; two, an increase in demand for our visual ad product, which has resulted in significant uplift in the eCPM and the expected year‐end launch of new visual ad products, where early betas have already seen great results; three, continued adoption of native advertising, such as sponsored listening, which is already pacing to exceed our revenue expectations for the year; and of course, a new high watermark for listener engagement.

Our strategy continues to be a focus on high‐quality, premium audio advertising, at ad loads that balance strong monetization with increased listener engagement. This allows us to achieve two goals simultaneously. Firstly, rising RPMs which directly drives expanded gross margins and profitability; and two, increasing total listener hours on the platform. We now stream more hours of music on mobile than YouTube streams hours of video in the U.S., and are second only to Facebook in total time spent. Ultimately, it is this massive scale that underpins our marketplace strategy.

Pandora now averages 24 hours per listener, per month, a new all‐time high, and total hours grew 7% year‐on‐year, or nearly 400 million hours, to 5.7 billion, also a new all‐time high. We view this as a very encouraging data point, particularly when you consider the number of free or heavily discounted on‐demand alternatives readily available to consumers. While total users remained essentially flat, we would expect both hours and users to grow over time as we launch new products, and as more people migrate to streaming radio and music services. Along with this migration, so too will dollars continue to shift from traditional radio budgets to digital platforms.

Already, elements of our broad music ecosystem strategy are beginning to take shape. Since Ticketfly has begun joint selling with Pandora, they have exceeded their bookings projections. Why? Because we are combining ticketing inventory supply with a massive and highly targetable audience of listeners. Venue owners see the power not only to drive ticket sales, but also to more effectively program based on precise knowledge of their customer base. These early results provide a glimpse into the future of live events in our marketplace.

The Uber partnership speaks to the audience growth prospects. Why did Uber put us in the driver's seat? We are the perfect solution for in‐car listening, a massive category that remains largely untapped with our current offerings. We are the service their drivers and riders use already. In just three weeks, more than 50,000 Uber drivers have used Pandora via the Uber integration, and average daily listening is over 2.5 hours per user. The early adoption numbers are even more encouraging given Uber is still in the process of rolling out the integration to their complete driver audience. Ubiquitous, habitual use is the foundation of our marketplace, and we are forging towards that vision.

A central part of our approach to a complete marketplace for music is providing a valuable platform for artists. This quarter we announced that we generated more than $2 billion in all‐time royalties paid for artists and songwriters, just nine months after hitting $1.5 billion royalties milestone. We are proactively working to ensure that as we launch new products, they are fully licensed, and they are being paid correctly, which was the motivation behind our recently announced partnership with Music Reports Incorporated, MRI.

Mike will provide more details on our partnership shortly, but it is of the utmost importance to us that we implement this open and transparent approach to rights management and set a new standard for the industry as we continue to evolve the platform. Given these many milestones and developments, we feel energized about the next era of Pandora for fans, music makers, and advertisers, and as we evolve, we will continue to always do the right thing for Pandora shareholders.

Before I wrap, I want to say a few words of thanks to Peter Chernin, who effective yesterday, concluded his tenure on our board. We are fortunate to have had Peter advising the company for the last five years. His contributions, guidance and passion for building enduring companies has helped Pandora navigate a dynamic landscape and position the company for success. Over the past 15 months, the company has added some of the world's foremost experts in music, technology and digital advertising to the Pandora board, including: Tim Leiweke from AEG; Roger Faxon from EMI; and Mickie Rosen and Tony Vinciquerra from Fox.

Now, over to Sara and Mike.

Sara Clemens - Chief Operating Officer

Thanks Tim. As I mentioned last quarter, we are operating with our deepest ever bench of product and engineering talent, enabling an accelerating stream of product launches for all three of our key constituencies: fans, music makers and advertisers. Core to realizing our music marketplace vision is delivery of features that are equally delightful for all three of these groups. By driving deeper engagement with listeners, we present more opportunity for music makers and advertisers to connect directly with their fans and customers. The data created by these engagements fuels further refinement of our product and campaign approaches, improving the marketplace as it expands.

Today, I will cover some highlights from Q2 and speak to the opportunity they represent for Pandora as we look forward to the release of our new services later this year. We're delighted to announce that Thumbprint Radio, which provides a mix of listeners' greatest hits and new music we are confident they will equally love, is now the number one station on Pandora. It achieved this status in less than six months, attracting over 14 million total listeners. Under 25 year olds are twice as likely to tune in to Thumbprint as the average listener, and we see a higher level of engagement, a greater level of thumb feedback, and fans spending more time with Pandora. This product is a preview of the kind of dynamic personalization that forms the basis of our next‐generation service.

This quarter, we launched New Music Stations; genre‐based stations which provide a platform for the discovery of emerging artists. In the first 24 hours, we had over 1 million stations created. These stations are powerful channels for up and coming artists in building a fan base. Artists and bands that are showcased have seen a 10% increase in station adds. This plays forward for artists, as station creates are an input to spinning across Pandora. With emerging artists, for every listener that creates a station, we spin the music to over 45 additional listeners across the service. New Music Stations are a great example of product innovation that benefits fans and artists equally.

We have been experimenting with new forms of content in addition to music with our Serial and This American Life podcast series. Combined, to‐date these stations had close to 10 million listeners stream an episode, with a total of over 20 million episodes streamed. These results confirm that Pandora has the unique ability to leverage its current strengths in music streaming to expand into the go‐to destination for discovery of a broad range of audio content.

Research confirms that listeners would like to hear more non‐music content, but are currently frustrated by discovery. Pandora's ability to surface relevant content to a scaled and engaged audience uniquely positions us to solve this problem. Additionally, listeners had confirmed that content offered by Pandora is viewed as more legitimate than unknown sources, increasing both their appetite and willingness to engage with new genres on the platform. We believe half of Pandora users are already consuming non‐music content weekly on alternate platforms, demonstrating strong, latent demand. And on the advertiser front, brands that sponsor non‐music content are perceived favorably, highlighting strong monetization prospects. We are now exploring scaling these efforts as part of our roadmap.

On the Music Maker front, we have continued momentum with Pandora AMP. Over 300 million Artist Audio Messages have now been heard by fans. And we continue to see engagement rates at the top end of social and paid media tools, demonstrating the efficacy of this unique service. In Q2, we launched our first automated Ticketfly integration, enabling all events booked via the Ticketfly Backstage platform to be seamlessly surfaced via the feed of the artist's fans on Pandora. We have been working with Ticketfly promoters and venues, and building integrated campaigns, with considerable success.

The most recent highlight is with an Australian artist, Flume. To promote his new album and accompanying world tour, Flume created a custom mix tape and launched a tour presale supported by audio message, email and push notification. Pandora sold over 25,000 tickets in three days, and 20% of house tickets for the entire tour. Results were particularly strong at the Ticketfly venue, Forest Hills Stadium, in New York where Pandora and Ticketfly jointly sold 92% of presale tickets. The strength of Pandora's promotional power, combined with Ticketfly's cloud live events management platform, drove Ticketfly's strongest booking quarter ever. Jam Productions, one of the largest independent promoters in the United States, joined Ticketfly in June, bolstering our position in the Chicago market, and maintaining our drumbeat of iconic venues converting to the Ticketfly platform.

In advertising, Pandora continues to build innovative experiences that delight both listeners and advertisers. This quarter, we announced the launch of our visual advertising experience, a responsive mobile display unit that activates a creative that automatically adjusts to the size of the phone's screen. This unit includes muted video, and is a new attention‐winning opportunity for advertisers on Pandora. After listeners interact with the visual ad, we've seen a doubling of the number of listeners who engage with a brand's landing page for more than 30 seconds. Across fans, music makers and advertisers, all of our product innovation has one thing in common: it leverages our core scale of radio listening to enable successful launch of new, profitable services. Nowhere will this be more apparent than with the launch of the first wave of our on‐demand product later this year.

I'll now turn the call over to Mike to discuss our financials.

Michael S. Herring - President & Chief Financial Officer

Thank you, Sara. As this quarter's activities demonstrate, we remain on track with our five‐year plan to scale our platform. This quarter gives us increased confidence, we will drive significant gross profits and incremental contribution margin as we respectively grew non‐GAAP gross profit and non‐GAAP adjusted EBITDA margin by six percentage points and 12 percentage points respectively. We continue to have a clear path to a 60% non‐GAAP gross margin and a 20% non‐GAAP operating margin by 2020 in our core radio business.

Driving our gross margin is our ability to expand RPMs versus LPMs, which will remain virtually flat through the next five years, adjusted for inflation. Total ad RPMs grew from $49.94 in the second quarter of 2015 to $53.34 in the second quarter of 2016, with total LPMs flat at approximately $32, thus the entire increase to RPM resulted in incremental gross margin. This is a key facet of our financial model that is unique to Pandora among industry leaders and speaks directly to our long term profit potential.

Looking to the long term, as we continue to execute against our music marketplace vision, we are actively building the requisite new products and securing the appropriate licensing. To that end, we announced a partnership in the second quarter with Music Reports Inc., or MRI. MRI enables us do two things: one, administer our direct deals with publishers; and two, scale licenses across the thousands of publishers in the industry. The license agreement we are offering through MRI is publicly available on their website for any publisher to opt in to. Within just a few weeks of launch, more than 1,800 publishers had signed up. One thing to remember is that this agreement only relates to publishing rights for compositions or the songwriter portion, so it doesn't impact our SoundExchange or label direct deals for sound recordings.

I'm proud that Pandora's open and transparent approach to rights management is setting a new standard, and is one that the music industry has widely embraced as a fair and improved way of doing things. We have also begun to secure the sound recording licenses necessary for the next phase of products we intend to launch, including more than two dozen licenses with independent labels already inked and significant progress ingesting the music catalogs necessary to execute on the back-end royalty administration of services licensed through direct deals.

We also continue to have constructive conversations with other independent labels as well as the three major labels and are optimistic that fair, win‐win deals can be negotiated. We believe we have the necessary assets to create a profitable on‐demand business: namely, the near‐term profitability of our core streaming radio business, our large existing audience, and our unique scale of music listening data. Pandora looks forward to partnering to bring new innovative products to fans and music makers that will make significant contributions to our company and to shareholder value, as well as to the music industry as a whole.

Now, let's walk through our second quarter results in more detail. Starting with revenue, we ended the second quarter of 2016 with record revenue of $343 million, an increase of 20% compared to $285.6 million in revenue for the same quarter last year. Excluding contributions from ticketing services, revenue was $320.3 million, an increase of 12% over the year‐ago quarter. Advertising revenue increased 15% in the second quarter of 2016 to $265.1 million, compared to $230.9 million in revenue in the same quarter last year. And we crossed the milestone of $1 billion in trailing twelve months total advertising revenue. Local advertising revenue, a strong engine of growth for Pandora, accounted for 28% of total advertising revenue in Q2.

Second quarter subscription and other revenue was $55.1 million, an increase of 1% over $54.6 million in the same period in 2015. Our end of period paid subscribers increased approximately 23,000 to 3.9 million, an increase of approximately 1% year‐over‐year. Ticketing revenue in the second quarter was $22.8 million as the Ticketfly business is solidly meeting expectations. Second quarter gross transaction value, excluding box office sales, $160 million, growing approximately 20% year‐over‐year. We transacted approximately 3.7 million tickets, excluding box office sales, in the quarter, which were purchased by approximately 1.6 million unique ticket buyers for approximately 38,000 live events, significant growth of approximately 20%, 20% and 30% year‐over‐year, respectively.

I'll note that while total revenue was slightly lower than our expectations for the quarter, we were able to make up the shortfall through expense controls, as reflected in our EBITDA, which I'll walk you through momentarily. Our focus this year is on investment in new products and solidifying the core business. Our ad sales continue to grow, without increasing the number of sales heads significantly and maintaining our focus on users, hours and growing engagement. Speaking to engagement, we saw an average listening of 24 hours per month, per user, which is two times that of the next largest service in the United States, as verified by third‐party measurement sources. Growing engagement provides additional hours for monetization opportunities and is a leading indicator of listener loyalty and retention.

The incremental revenue growth translated directly to gross profit and to EBITDA. During the second quarter, we had $45.7 million in sequential revenue growth and $36 million in non‐GAAP gross profit improvement despite hours' growth increasing content costs, improving gross margin by 630 basis points sequentially. We leveraged operating expenses by keeping expenses flat while growing revenue. As a percentage of revenue, non‐GAAP operating expenses decreased by approximately 600 basis points sequentially, resulting in $32.3 million in adjusted EBITDA improvement and an adjusted EBITDA margin improvement of 12 percentage points, which combined with flattening LPMs, provides clear evidence that as we improve monetization, we can see a clear path to profitability.

Moving on to EBITDA, consolidated adjusted EBITDA for the second quarter was a loss of $25.1 million, compared to a positive $16.3 million in the same quarter last year, and a $32.3 million improvement from the first quarter as 71% of the revenue increased sequentially flowed through to EBITDA. Adjusted EBITDA excludes $32.4 million in expense from stock‐based compensation, $14.4 million of depreciation and amortization expense, approximately $6 million in other expense, and approximately $1.5 million in benefit from income taxes.

Second quarter 2016 GAAP net loss per share was $0.33. Non‐GAAP basic and diluted net loss per share was $0.12, which excludes approximately $32.4 million in stock‐based compensation expense, approximately $5.1 million in amortization of intangibles, and approximately $1.3 million in amortization of non‐recoupable ticketing contract advances. GAAP and non‐GAAP basic and diluted EPS were based on 229.7 million weighted average shares outstanding.

Content costs represented 51% of total revenue in Q2, a reduction of approximately 600 basis points from Q1. As I mentioned earlier, our ability to drive leverage on these costs is dependent on our ability to increase RPMs in excess of our LPMs. Q2 2016 total RPMs reached a record second quarter high of $56.56, increasing by $2.65 or 5% compared to the year ago quarter. For the quarter, total LPMs increased by $6.64, or 27% compared to the same quarter last year, but grew only 0.5% compared to Q1. As a reminder, we expect to see significant LPM growth in 2016 versus 2015, due to the step up in royalty rates. However, over the next four years, LPMs will only grow at a rate equal to the CPI index, while RPMs will grow based on our ability to continue to improve the monetization of our service.

During the second quarter, non‐GAAP gross margins were 38%, compared to 48% in the year‐ago quarter, primarily the result of costs associated with content, as discussed previously. Sequentially, gross margins improved 630 basis points from Q1. Turning to operating expenses, we increased head count 34% year‐over‐year to 2,334 employees at the end of the second quarter of calendar year 2016, from 1,746 employees in the same period last year. For the second quarter of 2016, non‐GAAP sales and marketing expense was $105.7 million, or 31% of revenue, compared to $80.7 million, or 28% of revenue in the second quarter of 2015, as we continued to ramp our sales team to 508 QBSRs at the end of Q2, 154 of which were focused on local markets, and increased our brand and direct marketing activities.

Included in sales and marketing expense in the second quarter are commissions on subscriptions that we paid Google and Apple totaling $10.9 million, and $24.5 million in brand, direct response and SEM activities. The recent changes to Apple's in‐app subscription policies will result in savings to Pandora in future commissions due for subscriptions sold through their platform. In the second half of 2016, we expect to see savings of $2.5 million and $2.7 million in Q3 and Q4, respectively, but over the long term, these changes are beneficial to our business model as we launch and grow new subscription businesses. Non‐GAAP product development expense was $24.7 million for the second quarter, or 7% of revenue, an increase of 85% compared to $13.4 million in the second quarter of 2015, driven by significant organic investment in engineering resources and the acquisition of Next Big Sound and Ticketfly and the employees from Rdio in 2015.

As we have said previously, we believe product development is an investment in innovation to drive revenue 13 months to 36 months out, and thus we remain committed to increasing our spending in this critical area. Non‐GAAP G&A expense was $31.9 million or 9% of revenue, compared to $31.2 million in the same quarter last year or 11% of revenue. Turning to the balance sheet, Pandora ended the second quarter with $311.3 million in cash and investments compared to $382.5 million at the end of the prior quarter.

Cash used by operating activities was $45.5 million for the second quarter compared to $9.9 million of cash used by operating activities in the year‐ago quarter driven in part by the timing of routine working capital movements. Capital expenditures were $20.2 million in the second quarter, primarily driven by server equipment and office buildouts. Internal‐use software costs were $7.1 million in the second quarter, driven by capitalization of engineering costs associated with the development of the new subscription services.

Now, I'll wrap up with some thoughts regarding our guidance for the calendar year 2016 and the third quarter. First, regarding calendar year 2016, due to softness in national advertising in the second quarter and the uncertainty around political spending, we are lowering our full year revenue outlook to a range of $1.385 billion to $1.405 billion, or year‐over‐year growth at the mid‐point of approximately 20%.

We expect calendar year 2016 adjusted EBITDA loss to be in the range of $70 million to $50 million, consistent with our prior guidance. Adjusted EBITDA excludes forecasted stock‐based compensation expense of approximately $142 million, depreciation and amortization expense of approximately $62 million, and a benefit from income taxes of approximately $0.3 million, and assumes minimal cash taxes, given our net loss position for the year. Basic shares outstanding for the calendar year 2016 are expected to be approximately 231 million. We are also forecasting a non‐GAAP effective tax rate between 30% and 35% cumulatively for the year.

For the third quarter of 2016, we expect total revenues in the range of $360 million to $370 million, achieving year‐over‐year growth at the mid‐point of 17%. With LPMs approximately flat quarter‐over‐quarter, we expect the sequential quarterly [technical difficulty] (28:10) flow directly to gross profit and materially to the bottom line, and thus adjusted EBITDA for the quarter will improve sequentially to a range of a loss of $5 million to a profit of $5 million for the third quarter. Adjusted EBITDA excludes forecasted stock‐based compensation expense of approximately $35 million, depreciation and amortization expense of approximately $16 million, and a provision for income taxes of approximately $500,000, and assumes minimal cash taxes. Basic shares outstanding for the third quarter of 2016 are expected to be approximately 233 million.

In summary, we are on track with our long‐term plan and executing against our strategy. The trajectory of the business in the second quarter gives us further confidence that we have the right plan in place to realize our vision and create shareholder value, and we remain excited about the opportunities and our future as we evolve our platform and build a massive marketplace for music.

And with that, we're ready to take some questions. Operator?

Question-and-Answer Session

Operator

Your first question is from Heath Terry from Goldman Sachs.

Heath Terry - Goldman Sachs & Co.

Great. Thanks. Mike, can you unpack for us the difference between listener hour growth accelerating and revenue growth decelerating? On an individual basis, what are the metrics that make up, that doing whether it's sell-through, ad load, pricing, just want to get a better understanding there since we've started to see this stronger listener hour growth after the last couple of years? And then, Tim, would certainly appreciate if you could just walk us through a little bit more of what it means to be signing the independent deals that you're signing. Are you getting what you're looking for in those deals in terms of the kind of the flexibility to offer different price points that you've asked for? What kind of progress you've made so far informing the discussions you're having with the major labels?

Timothy Westergren - Chief Executive Officer & Director

(30:22) and I'll follow up.

Michael S. Herring - President & Chief Financial Officer

Yes. So hours growth still is – while it's rebounding and starting to accelerate and we're excited to see that. It's still significantly below our revenue growth rates, and that's in part because we continue to drive our efficiency or RPMs up. But there is a correlation between these, because although we are driving to certain revenue targets and those revenue targets are critical in terms of driving profit margins and funding the business, we're also balancing that against long-term users and hours and engagement goals.

And so there is the ability to actually both drive revenue growth and drive hours growth at the same time is a balancing act that we play as a company over time. And we're looking at a long-term value optimization of balance between the ad-supported business and having a large population in order to fuel other businesses, whether it be subscription services or live events promotion or ticketing. That balance is something that we try to strike over time. So we're not overly concerned about optimizing one versus the other. It's about optimizing those metrics as a portfolio.

Timothy Westergren - Chief Executive Officer & Director

And just to add, you have to put that in the context of what's ahead, right. So our strategy in front of us it to offer this broader product, and we want to have our audience as large as possible when we hit that and bring that to market. So we are trying to balance this monetization efficiency with maximum retention and engagement. So we've been doing that for a while now.

And to your question on the deal front, absolutely so far we've been able to get deals that allow us to offer the product we want. It's one of the really encouraging parts of this so far. I think that what we're finding is that the case we're putting in front of the rights holders, people are believing that we are bringing something that is truly incremental and that is not just going to drive the high-end subscription but will also bring in a big part of the market that is not currently paying. And Pandora is uniquely positioned to do that.

So we have this enormous audience, approximately 100 million people every three months, and our ability to introduce them to these products is unique. And so our confidence just keeps growing about being able to not only get the feature set we want, but also the economics we want long-term. And we won't sign deals that don't preserve that. That's the beauty of having actually a healthy core business to start with.

Operator

Your next question is from Ben Swinburne from Morgan Stanley.

Benjamin Daniel Swinburne - Morgan Stanley & Co. LLC

Thank you. Tim and/or Sara, on the engagement front and the hours growth, you guys have been doing a lot behind the scenes on the product side. You talked about Thumbprint. You also talked about balancing ad load. Could you just maybe spend a minute on what you think were the biggest drivers in the quarter and really year-to-date in the accelerating engagement levels we're seeing just so we can think about sustainability?

And then, Mike, on the cost side, I think, you guys had talked about $120 million of costs associated with on-demand this year, and as you mentioned, you've put in some cost controls. Can you just talk about what you're spending at this point for the year on the new initiatives? And if any of that money moved into the cash flow statement, so it's still showing up just in another part of the financials that would be helpful color? Thank you.

Michael S. Herring - President & Chief Financial Officer

So in terms of engagement, there are lots of pieces that drive engagements. At the cornerstone of that is making the product better from a playlisting standpoint. I think, you should never forget the intellectual property capital that goes into actually selecting songs, which Pandora is unique in. We have an enormous data science team now that's working on that and running experiments daily with trying different algorithms, and we just keep optimizing that, and that's sort of a rising tide across all of our listenership. You've got special products like Thumbprint and promotional programs, and you've got also the growing distribution footprints into more places, more cars, more CE devices. We keep having these new products come to market. You add that altogether that drives engagement. But Sara you can give some more color on that.

Sara Clemens - Chief Operating Officer

Yes. So, certainly, the core algorithmic improvements underpin a steady year-on-year increase in hours. I think what you've seen in the last six months is this accelerating level of product improvement on top of that. So the launch of the services that I described, Thumbprint, the podcasting, new music stations. And you can expect to see us continue to develop new features like that, much in the same way that we did with podcasting, to try it out, understand our audience appetite, and then really look at scaling it once we have some understanding of what it is that our listeners are looking for and the types of services that we can provide the content creators.

I think the other thing to note here is over the course of the last year, we have run hundreds of campaigns with Music Makers. So these are leveraging the Pandora AMP promotional tools. But many of these drive to content experiences. So a great example of this is what we call the ATL station, which is an Atlanta local station that we created. We did this last year. It's a hip hop station. And we've been building a really strong audience for that, which we then created a live event around, which we ran in Atlanta. We did live streaming of that, which we had close to a 100,000 people tune into the live stream and a 1 million people engage with the station. And we've just had superstar rapper Gucci Mane who's chosen to collaborate with us to do a takeover of the station to celebrate the launch of his new album. And so what you're seeing here is artists really being one of the key promotional and marketing tools for us to drive additional engagement with fans and that all increases hours. So I think, again, as we scale the Music Maker Pandora AMP platform, you will see improved benefits from that as well.

Michael S. Herring - President & Chief Financial Officer

And just really, quickly on your question. Yes, we talk about certain capitalized software costs associated with new subscription services. If you watch our cash flow statement, that increased from about $2 million to $3 million a quarter at the end of last year to about $7 million to $8 million a quarter – low $7 million this year. That difference, which is about $3 million to $4 million a quarter, is related to subscription services and certainly is part of that $120 million investments.

Operator

The next question is from Stan Meyers from Piper Jaffray.

Stan X. Meyers - Piper Jaffray & Co. (Broker)

Thanks, guys. Just wanted to get a little clarity and guidance on the marketing spend of the on-demand product, I know you guys mentioned in the past about $40 million. Just based on the guidance, where is that $40 million sitting across the quarters? And secondly, maybe you can talk a little bit about the I guess Apple's submission to the CRB on interactive rates for 2018 through 2020 and how that may impact your economics and negotiations? Thanks.

Michael S. Herring - President & Chief Financial Officer

So the marketing dollars, we have it planned to the back half of the year. It is more heavily weighted to Q4 based upon our hope to launch new products in this year. But we do plan to spend marketing dollars relatively significantly in Q3 as well leading up to that launch assuming everything is on plan. Tim, do you want to comment on Apple?

Timothy Westergren - Chief Executive Officer & Director

Yes. So these are two distinct things. Apple is proposing something that they think works for them in particular. We're working directly with rights holders to find the right solution for the product we're offering. It varies across products. And again just to reiterate, I think, because we come to this as a core profitable product ad supported that has this high engagement and allows us to enter the market in a different way and it allows us to offer economics that are healthy for us but also healthy for the business. So we're not asking the business to solve our business model problems. But Apple's proposal – we don't think there's a one-size-fits-all business. So we're doing our deals directly.

Michael S. Herring - President & Chief Financial Officer

Yes, and let me just clarify a little bit for those of you who aren't as familiar with Apple's proposal. It's related to the publishing portion of the royalties only. It's not related to the sound recording piece. And so their proposal is basically they'll have a flat fee per stream for publishing royalties across all music plays, whether it's ad supported or subscription supported, which might work well for their purposes, but does create bump in a market where there's multiple businesses models and multiple approaches to delivering music to consumers.

Operator

The next question is from Douglas Anmuth from JPMorgan.

Douglas T. Anmuth - JPMorgan Securities LLC

Great. Thanks for taking the question. I just wanted to circle back on the advertising business in 2Q and Mike it sounds like you talked about some weakness in demand on the national entertainment and telco side and the lag in hours perhaps not being able to catch up to it. I would think that's where maybe programmatic would kick in more. So maybe you can talk about that. And also just trying to understand, if the ad load actually changed here during the quarter just as you're trying to ensure the highest quality experience? Thanks.

Timothy Westergren - Chief Executive Officer & Director

I'm just going to kick it off and pass to Mike, Doug. It's a great question. And we've been in this space for a while and we've developed a really good understanding where the pockets of demand are. And right now the money comes from really direct premium sold advertising. Programmatic will have its place over time, and when it's significant, we will be there. And we'll have a unique offering and we'll make the market, but the business is really, in audio that is. And we made the shift a year ago when we developed programmatic in mobile display, which has been a really great new product for us growing healthily inside the business. So we've been at this market for long and understand those things. That's not the quick fix for this.

Michael S. Herring - President & Chief Financial Officer

And what was your second question again, Doug? Sorry.

Timothy Westergren - Chief Executive Officer & Director

Is that it?

Michael S. Herring - President & Chief Financial Officer

Okay. Mike, next question please.

Operator

And the next question is from Mark Mahaney from RBC Capital Markets.

Mark Mahaney - RBC Capital Markets LLC

Great, thanks. Two questions; one, any color on the sequential decline you saw on active listeners? Is that largely in line with expectations? Thoughts about that going forward, or when you want to start really turning on marketing to kind of grow that? Are you just kind of timing that for the on-demand service?

And then secondly, any comment at all on political advertising and whether that's coming along as you would have expected. Do you think that there will be a fair amount of that coming your way as we get closer to November? Thank you.

Timothy Westergren - Chief Executive Officer & Director

What's the first question, again? Yes. So, good question, Mark. This is pretty much where we expected to be, more or less flat and I think what you said is right. We're timing this to hit the market, optimizing for monetization against growth, so that when we do bring new products to market, we're in a perfect position to be there. So, that's how we think about it. And then in terms of political, it's been an odd political season so far. We've got part of what we – that's factored into our second half projections. And we'll watch it evolve, and hopefully, especially on the Republican side, the dollar starts flowing more than they have been, so.

Operator

Your next question is from Jason Helfstein from Oppenheimer.

Jason Helfstein - Oppenheimer & Co., Inc. (Broker)

Thanks. I just want to follow-up to the Doug's question. Then I have my own question. I think there was a question about, it sounds like there was a bit less volume in the quarter, but then you talked about a bit of weakness in national entertainment and telco. Were they connected, I guess, I think that's the question. So was there a lower ad load because you chose to drive engagement, or was it a function of those weaker categories and are those categories you're seeing the same thing in the third quarter?

And then to the extent if you can talked about the discussions or the industry right now, when you think about competition, if your competitors are still growing at the $10 a month price point, talk about what's the urgency of the labels to grant you a right to sell a different type of service at $7 or $8 or some of the things that have been talked about. So any color would be helpful there? Thanks.

Michael S. Herring - President & Chief Financial Officer

Well, thank you for bringing Doug's questions back up again. Yes, so we did not reduce ad load. In fact, the maximum ad load stayed around six to seven ads per hour from an audio perspective. It's been consistently that really since the mid-2014, so quite some time. Now the number of the effective ads delivered was about 3.3 in the quarter. So up year-over-year and it's certainly up sequentially significantly. So it's not that we're serving a lot different ads. We are balancing that against user and hour growth and optimizing for engagement. Like I said, engagements are the number one indicators for retention. You want to focus that in terms of how you're developing loyalty in your users. And if you're bringing new products to market, it's an important time to optimize that.

So we aren't pushing aggressive ad load because we are balancing it against other pieces, but it's not that we're holding it back artificially either. We're trying to balancing at the right mix between driving the right revenue, driving the right leverage in the business and setting this up for a long-term success.

In terms of the industry's view, competitors drawing a $10 price point, first of all, let's just take that as an assumption. It's probably not super correct. If you look at what competitors are actually growing at, if that discount is pretty significantly off that $10 price point, you're seeing the average revenue much, much less than that number from those who disclose it.

So why would the labels want to let Pandora grow at that price point or a lower price point, it's because we can access a much bigger audience with a more differentiated product that can drive incremental subscribers into the marketplace generally.

And because, frankly, we have a large ad supported, profitable business that we can use to up-sell and cross-sell users into that opportunity, we can actually have a lot more price discipline (46:16) than anybody else who's reliant solely on thinking about it as a funnel into subscription. You can look at it as one piece of the puzzle, call it a product in a portfolio that you're optimizing against.

That approach is resonating extremely well in the music industry, where rather than growing through price cuts or through discounts, you grow through providing value propositions that are unique to listeners and get those listeners to up-sell and cross-sell into new products, because they want to and they want to pay for it. And that's resonating really well. So we think that that story will drive the true partnerships with music industry, that win-win deal that I was talking about.

Timothy Westergren - Chief Executive Officer & Director

I'd also add to that. One of the issues of current services is churn and you can drive new subs, but how long can you keep is a big question. And one of the strongest correlation to churn is engagement. So you need to develop a product that actually engages listeners and causes them to believe it's value we're paying for. And I think labels are seeing we have that ability too, when you look at our existing product and how successful we've been at driving engagements, we bring that same expertise to this domain. And I think we'll have the ability to not only offer these products but through that lower churn and better economics ultimately.

Operator

Your next question is from Sameet Sinha from B. Riley.

Sameet Sinha - B. Riley & Co. LLC

First is any of these promotional tools that you highlight every quarter between artist communication with listeners, is that taking away from monetizable listener hours? Is that part of the balancing act? And secondly, local advertising seems to have slowed significantly from Q1. And I also saw that local sales people were flat quarter-over-quarter. Is that a direct correlation? And can you talk about how you think about that increased hiring or keeping it at the same level? Thank you.

Sara Clemens - Chief Operating Officer

This is Sara. I'll take the first question on the Music Maker product. So one of the delightful things about our Pandora AMP tools is that listeners love to engage with what are effectively expressions of love from artists towards them on using these unique promotional capabilities that we've built. And we tested these very carefully as we launched them. So you'll recall, we launched Pandora AMP in October 2014. Initially with an insight to tool and the insight of rolling out the audio messages and other tools. And one of things that we were testing for was whether or not listeners receive these messages and pieces of content as advertisements. And therefore, to your point, they might potentially displace advertising inventory.

It's absolutely not the case. Listeners could see these as content. So they see them as an augmentation to the listening experience. And so we are able to do these in addition to the advertising loads and in that respect, they're actually beneficial, because they reduce our total cost of content to the extent that they are taking up time within the stream.

Michael S. Herring - President & Chief Financial Officer

Yes. And just to answer your question about local ad revenue, actually it stayed at 28% with that sequential growth of total revenue is a testament to the strength of local ad revenue, it's not a knock. Local ad revenue doesn't have the seasonality of national and thus Q1 should have been generally a high watermark of their overall percentage of ad revenues, so as we move into Q2, we had a big jump in overall revenue, jump in ad revenue, local revenue sequentially grew equally with the rest of the market, despite not normally being seasonal. And so it had a very strong quarter.

It is growing off of – in terms of growth rates against prior years, it's a 300% increase over the last two years. So it's growing off a small base since we've invested. Your point that we haven't added local ad people as much as we have historically, it's really an internal decision to focus on optimizing the business versus investing aggressively in growth, while we're using those investment dollars to bring new products to market.

We still are big believers that local ad revenue over the long-term is a big driver of the $2.4 billion target in 2020. But that takes investment and we need to balance when we're applying those investments at different times in order to drive that. And right now, we're really driving the optimization of the people in the field today in order to fuel the rest of the business.

Operator

The next question is from John Egbert from Stifel.

John Egbert - Stifel, Nicolaus & Co., Inc.

Great. Thanks. Could you talk a little bit more about the segments of your user base that are driving the upside to listener hours? Is it coming from like autos or TVs or consumer electronics devices where your monetization efforts are a little more nascent?

And on the user side, given your strength in listener hours, it seems like there's more going on than what your monthly active listener count tells us alone. Can you give us any directional guidance on how daily or weekly users are trending? Are they growing faster than the monthly listeners? Thanks.

Michael S. Herring - President & Chief Financial Officer

Yes. So that's a good point. As CE and auto have grown significantly, and they're growing much, much faster in terms of hours, again, off lower basis, than mobile audiences are for example. And so although they are small percentages of our overall hours relatively, I think together they make up about 10% of our total hours today. They're growing quickly, but they're not necessarily about driving users. It's about getting users to use Pandora more. And so it's driving hours from the same users and just accessing their – becoming a music habit in different parts of their day, whether it's driving in the car, or it's when they come home and they're making dinner, they're flipping on their Sonos box, their Roku box. It's adding these multiple touch points in the day that helps us build engagement that Tim mentioned earlier. It also builds our overall hours position, which is a sign of – and it builds loyalty, it builds retention in those users. So all those things are good.

Other terms, you mentioned the monthly active users. We like talk about the 100 million users that come every 90 days or so. But internally when we're looking at it, what we probably look at more than anything else is daily active users, and it's not a metric that we release externally, but it's a big one how we measure it and in part because a lot of initiatives that touch people whether it's in the car, or in CE, it's getting them to use it on a more frequent basis, if they're work users using it at home on the weekends, et cetera, and that is up significantly, a much, much higher daily active user growth year-over-year; very different patterns than if you look at the industry standard MAU.

Sara Clemens - Chief Operating Officer

Just one other point there. In terms of some of the opportunity around non-music content, we're seeing this against our comedy content that we have on the service today and then the podcasting. We're seeing that as incremental hours because it's actually being consumed at different hours to pick music consumption hours. So as we actually build out some of these new products, we're also seeing them drive incremental upside.

Operator

The next question is from Nat Schindler from BoA.

Nat H. Schindler - Bank of America - Merrill Lynch

Yes, hello.

Timothy Westergren - Chief Executive Officer & Director

Hi, Nat.

Nat H. Schindler - Bank of America - Merrill Lynch

Is this me? It didn't sound like me and my name. But quickly, can you help me out on telling us what happened a year ago with the skip changes you made to the algorithm and how that affected hours one year ago? And were those changes kind of a step down in hours that you took or did they affect things over time?

Michael S. Herring - President & Chief Financial Officer

So those changes occurred really starting in March of last year through May. So, Q2 is kind of the first time we're lapping those changes which reduced skip. To those of you who are not familiar with this dynamic, as skips are reduced through play-listing or artificially and in the case we're talking here is, a significant improvement in play-listing technology that drove the average numbers of skips down materially, which meant that completed songs played longer. When skips occur, the recorded hours are higher even though listening is less, because even if you listen to one second and skip, we report the full three minutes, because that's what we paid for, the full four minutes.

As skips reduce, it's a better sign of engagement, but reported hours actually decline or have a headwind, maybe decline is too strong of a word. And if this significant change that was rolled out in the first half of last year had a headwind against hours last year, that was several hundred basis points.

Coming into this year now, when we compare this year to last year, we're talking more apples-to-apples, so with a step function rather than something that gradually continues to increase. I mean, I think we are always working on improving play-listing and such, so we expect over time things like skips to decrease on average, but it was rare to have something so material where you would see several hundred basis points of headwind in the growth.

But if you look at Q2 to Q2 and certainly for the rest of the year, it's going to be comparing apples to apples, so the 7% growth year-over-year in Q2 from hours, that's an apples to apples within degrees of reasonableness comparison of hours, and exceeded our expectation certainly.

Operator

The next question is from Robert Peck from SunTrust.

Matthew C. Thornton - SunTrust Robinson Humphrey, Inc.

Yeah. Hey, guys. It's actually Matt in for Bob. Couple of questions if I could. You talked about entertainment and telco verticals being weak particularly at the national level and you talked a little bit about political ad spend maybe a little bit below plan. I'm wondering how you're thinking about the Olympics in 3Q, if that's any type of a headwind and if that's contemplated in the guidance for 3Q and the back half of the year?

And then, just jumping over to the new products that you're pushing towards, it sounds like the timeline is still, if I heard you right, Mike, it sounds like the timeline is still pushing towards 4Q. But I'm curious if you're still expecting or what the latest thinking is around maybe cash payments upfront to kind of cement some of those deals. I think that was part of the reason for the convert raised late last year. So I'm just kind of curious what the latest thinking is there as well? Thanks.

Timothy Westergren - Chief Executive Officer & Director

I will just hit the Olympics question quickly, and that's one piece, certain advertisers might focus on it, but I wouldn't say it's reflected in guidance in some material way, it's just one of hundreds and hundreds of factors that we look at in the next six months.

Michael S. Herring - President & Chief Financial Officer

And so, deals are still being negotiated, so I can't speak directly to what's actually going to happen yet because it's unknown. We absolutely are committed to and believe we're optimistic we'll achieve a timeline where we can launch products by the end of this year. It's hard to say exactly what month that might occur or whether there is upfront payment associated with it. But I would say though that we did fortify our balance sheet last year in order to be able to negotiate from a position of strength, meaning that whether it was from upfront payments of some kind or just being able to invest appropriately in the opportunity in launching a product, we wanted to be able to make sure that we weren't restricted by capital issues, and we feel like we're in a very comfortable position from a balance sheet perspective.

Timothy Westergren - Chief Executive Officer & Director

And reiterate, in terms of as we think about these deals, we are doing deals that work for us, that preserve our long-term economics, and because we have a profitable core business, we're able to do that. It's very important for us.

Operator

The next question is from Mark Kelley from Citigroup.

Mark Kelley - Citigroup Global Markets, Inc. (Broker)

Hey guys. Thanks for taking the question. I know you can't talk a ton about the negotiations with the labels, but just curious if you get the sense that our revenue share agreement makes more sense relative to your per stream rate or how we should think about that? And then second, can you remind us what the monetization level is between local and national, anything to help us think about how RPM is different between the two? Thanks.

Timothy Westergren - Chief Executive Officer & Director

Yeah. Sure. So, it's all relative, the percent of revenue versus the per stream really depends on how big the per stream is or how high the percentage of revenue or how low either is. I think what's more important than one versus the other is that it's one-dimensional or there is a clear path to leverage for the distribution for Pandora. When you have multiple terms that eliminate the possibility for the service to actually leverage cost in a meaningful way, because if they leverage one direction, they are caught by one term, if they leverage in another direct, they are caught by another term, that they are creating that, that's a lose-lose situation and that's obviously not going to build a partnership, it's going to build any value long-term.

And that's I think something that we have stuck with, it was a big part of our CRB argument is that getting into just a one-dimensional licensing structure, which the CRB judges agreed to, with their $0.0017 and essentially flat for five years, CPI adjusted from there. While a step up from what we were paying, it was a one-dimensional clean royalty structure that allows a company like Pandora to invest aggressively and understand what its return profile is. And I think once that's understood by the music industry, that that's how you drive investment, and it raises all boats, I think that's when the right deals come out the other side.

Operator

The next question is from Michael Graham from Canaccord.

Michael Graham - Canaccord Genuity, Inc.

Thank you. I'm just wondering, what kind of lead time do you need from when you are able to get deals signed to launching the service like, could we see a scenario where the deals get – negotiations get pushed to like mid-December hypothetically and you're still able to kind of hit your Q4 outlook? And then, can you just remind us on the international on-demand outlook, you had started out a year ago or so saying that you wanted to kind of do international and domestic all at once and then you focused more on domestic. Just remind us, was that more because of the complexity getting the deals done globally? Or was it more just sort of operationally managing expenses, that sort of thing? Thanks.

Michael S. Herring - President & Chief Financial Officer

So, I'll start by answering the second part of the question from Mark. So the monetization, that was for national and local, national was pretty consistent around – I mean, there's big network deals you can do at CPMs that are pretty low, like $4 to $6. We don't really do any of those deals and that's a part of a large upfront that involves spot purchases as well. The national spot rates that we realize here are more like $8 to $10 or $11 CPMs or higher.

And the local, because of the targeting aspect, drive significant value from that being able to reach a specific audience, can be anywhere from 50% to 200%, 300% national CPM. So, the investments on bringing that mix up, helps us drive higher RPMs in a blended fashion. In terms of lead time for deals, they tend to be signed within 24 hours to 48 hours of the product launch. We are building that product today, that's why we've been building it. We talk often that it's a race between the licensing team at Pandora and the product team in Pandora. Licensing trying to get the products under a complete full licensing, so we can bring them to market and the product team to build the products to be ready for launch and there is a chance one could beat the other, but I think they're actually on track to dovetail nicely, based upon our outlook today. In terms of international versus U.S., I'll let Sara answer that question.

Sara Clemens - Chief Operating Officer

So just thinking about international. So we remain confident about our prospects internationally, the experience we've had in Australia and New Zealand demonstrates demand for the service and uniqueness of the service. We are focused on rolling out the new on-demand product in our existing markets first, but absolutely see global as part of our path to 2020 and we'll be working with the industry on those licensing deals going forward.

Operator

The next question is from Barton Crockett from FBR.

Barton Crockett - FBR Capital Markets & Co.

Okay. Thanks for taking the question. I guess a couple of things if I could. First, just to understand, you guys were talking coming out of the fourth quarter that your core radio would get around $50 million of EBITDA this year and that in tandem with the $120 million of investment in the on-demand stuff, I just want to make sure that that you guys still see that type of trajectory for the core radio business?

And then in terms of the variance here on the softness in the entertainment and software and maybe political versus what you were thinking earlier in the year. Is this just kind of part of the lumpiness of the business or is there any kind of reach you could have through to maybe the macro conditions being softer or anything that is more concerning or is it just kind of a one-time kind of mishap?

Timothy Westergren - Chief Executive Officer & Director

Yeah. I wouldn't say it's a general concern for us. Like you said, there is some fluctuation in the business, like this is large and is complex. We are off by a couple of percentage points. Now that said, we don't like to miss and we were short of our expectations and we're attacking that immediately. The last quarter, we outperformed. So there is going to be some of that, some lumpiness, but it doesn't reflect any systemic issue for the business and we're still on track to connecting those larger numbers you quoted earlier.

Operator

The next question is from Ron Josey from JMP Securities.

Ronald V. Josey - JMP Securities LLC

Great. Thanks for taking the question. I was hoping, Mike, maybe you can provide us a little more insight on RPMs this quarter just given subscription RPMs I believe declined, while advertising RPMs slowed and I know we talked a lot about that advertising with some verticals being weak, but can you talk about maybe desktop and mobile there, given the change in reporting and if that's maybe a reason why advertising RPMs slowed? Thank you.

Michael S. Herring - President & Chief Financial Officer

Well, so I mean advertising RPMs continued to increase. They're slowing in terms of growth. As explained, it's not about, we're focusing on doing the right revenue dollars and we're also focusing on driving as many hours as possible. So, where in historically we might have limited hours more aggressively, we're not doing that as much now and you're seeing that that denominator grow pretty significantly, it's 5.7 billion hours in the quarter. The denominator of that RPM is growing pretty healthily.

While revenue continues to grow healthily, if the RPMs are going to step-up, we're not trying to get that as high as possible. In that, we're trying to optimize it for the opportunity across multiple dimensions. From a subscription perspective, those RPMs are staying very consistent, so that is pretty consistent. There are some other – it's subscription revenue and other and there are some other revenue that was in there last year, that is not in there this year and that's not usually material, but it does probably make it look like the RPM is down a little bit on a year-over-year basis. But essentially when we look at subscription unit economics, those RPMs are very consistent year-over-year.

Operator

Your next question is from Rich Tullo from Albert Fried & Company.

Rich R. Tullo - Albert Fried & Co. LLC

Hey guys. Thank you for taking my call, long call. What percentage of your revenue is derived from ad agencies versus advertiser direct?

Michael S. Herring - President & Chief Financial Officer

Well, it's actually a difficult question to answer because there is a lot of sort of hybrid models out there today in our sales model. But it's probably only about 30% to 40% of our ad revenue comes from – is agency assisted and the rest of it is sold direct to the customer itself.

Operator

Your next and final question is from Peter Stabler from Wells Fargo Securities.

Peter C. Stabler - Wells Fargo Securities LLC

Thanks very much for taking the question. So I wanted to ask a little bit about the RPMs again. And Mike, can you give us a sense of how demand is distributed across the markets where you have local sales force on the ground. So if we think about the early cohort market, San Francisco, LA, DC, Boston, et cetera, what we would view as high demand markets. How does sell-through look – what does sell-through look like in those top early markets versus the second, third cohorts? And just through that could you help us just understand a little bit of the drivers behind RPM, because it sounds like max ad load was not lifted, so it sounds like sell-through. So I guess I'm just trying to understand where the sell-through gains were, were they in top markets, were they in lower tier markets? Thanks so much.

Michael S. Herring - President & Chief Financial Officer

So, in terms of driving RPM year-over-year with a max, max ad load hasn't changed and there's kind of two primary things there, one is exactly what you're talking about, the growth of local as a percentage of ad revenue, but also in total dollars, that's in primary markets where we are – have been in local for a long time, local media can be as high as 60%, 70% of the audio revenue in that market. Those RPMs are going to be running even though we're maybe hitting a maximum of 5.5 ads to six 6 ads per hour, RPMs now there in high value demos are well north of 80%, 90% even triple-digit RPM.

And that's because you're getting a high mix of local audio ads, but you're also seeing this year versus last year, much better mobile display monetization, and that's a big factor that's been driven by the launch of our mobile programmatic platform last year, and it's still building momentum, it's still relatively small, but it has brought sell through rate up pretty significantly and brought the floor CPM up significantly for mobile display. And that's allowing us to increase RPMs and increase monetization of hours year-over-year without pushing aggressively on audio ad load side of things.

Dominic Paschel - VP, Corporate Finance & Investor Relations

Great. With that we'll go ahead and close out today's call. If you happen to be in Miami tonight, you can join us with Passion Pit with the Lexus Pop-Up concert series. Next week we'll be with Matt and Kim to kick off Lollapalooza on July 27 in Chicago. That will be sponsored by Boston Beer, Sobieski Vodka, and Ashley Furniture. And finally if you want to join us for a big summer crush back-to-school concert, which we announced today with 5 Seconds of Summer, Fergie and Daya, sponsored by Amazon Prime, Best Buy, and many others, please shoot us a note at investor@pandora.com and we'll get you added to our guest list.

Mike, can you takes us back to the Toyota (1:11:15) please.

Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.

THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.

If you have any additional questions about our online transcripts, please contact us at: transcripts@seekingalpha.com. Thank you!