Jennifer Witz is the CEO of SiriusXM, a quietly fascinating company. You probably know it as the satellite radio brand that’s in virtually every new car, but it also owns Pandora, a huge podcast network that includes 99% Invisible and Team Coco, and a content operation that supports huge stars like Howard Stern and has deals with every major sports league.
What’s particularly interesting about SiriusXM is that it started as a distribution play — actually, two distribution plays. Sirius and XM were two different companies. They both launched competing satellite radio services in the early 2000s, basically pre-internet, and eventually merged.
Sirius is effectively the dominant market leader for built-in premium audio in cars. At the same time, the competition is here. Spotify exists, podcasts exist, and yes, the SiriusXM app exists as well. And as the infotainment system in cars gets ever more complex and computer-like, the Sirius experience has to keep up. A thesis of Decoder is that distribution affects what kind of content you make, so I wanted to know how Jennifer is thinking about all of those different kinds of distribution. What are the differences between what Sirius makes for podcasts, for Pandora, and for the satellite system?
On top of that, the state of car software is in total flux. GM just announced it won’t support CarPlay in new EVs. Lots of companies are using various riffs on Android. Something called QNX keeps bubbling along. Tesla has its own thing. It’s a mess. And Sirius has to support all of it with applications that compete with Big Tech companies, all while continuing to integrate the satellite hardware into the cars themselves — on top of putting satellites on SpaceX rockets.
That’s a lot for Jennifer to manage, and we covered a lot of ground. I also think I got Jennifer to agree to move Road Trip Radio to the Sirius channels, so for those who know, that’s a win.
Okay, Jennifer Witz, CEO of SiriusXM. Here we go.
This transcript has been lightly edited for clarity.
Jennifer Witz, you are the CEO of SiriusXM. Welcome to Decoder.
Thank you, Nilay. It’s great to be here.
I am very excited to have you here. I’m a Sirius customer, so I just have a number of feature requests. That’s it. That’s what we’re doing for the full hour. My wife is a hardcore user of the app; she has a number of feature requests, so get ready.
Great, I’d love to hear them.
Let’s start at the very beginning. One of the themes we talk about in Decoder all the time is that your distribution often affects what you make, and we see this over and over and over again. I’ve explained it a million times. YouTube is my favorite example. We all know what a YouTuber looks like. Sirius is a really interesting company because it has so many different kinds of distribution, which are kind of the product. It started as two different satellite companies with competing ideas about distribution. You have Pandora, which is an internet radio company. You have Stitcher, which is a podcasting bet across all kinds of other people’s distributions. How do you think about that?
It really starts with Sirius and XM, which came together about 14–15 years ago now. That is the primary source of our revenue. Subscription revenue from SiriusXM represents about 70 percent of our overall revenue, so we’re really focused on how we can continue to deliver value to our SiriusXM subscribers. And the bulk of those subscribers are listening to us over satellite in the car but it really limited our ability to do certain types of distribution.
So the content that we provide over satellite is really broadcast, right? One to many. The pipe is only so big; you can only put so many channels on the air. But increasingly, with our new platform that’s rolling out, which we call 360L, we can have a lot more interactivity in our service, whether it’s just recommendations or more content in general but as you mentioned, we have much broader distribution. If you start at the broadest point, it’s what we’re doing with podcasting, and that’s we believe in podcasters and content creators who want to distribute their content broadly across all third-party platforms. And so we have really strong business there as well, and that helps feed into SiriusXM over time.
I want to start with the satellite business. Like I said, I’m a customer, so I have a brand-new car that has 360L in it, and you can’t tell the difference between when it’s using mobile broadband and when it’s using the satellite service. Then, I have a car from 2016, which actually only has a Sirius radio. It doesn’t have an XM radio. And this, to me, is one of the most interesting technology dilemmas that maybe any CEO has to face. Sirius launched in 2002, XM in 2001. I remember the XM launch and that the first two satellites were named Rock and Roll. That’s stuck in my memory. You have to launch satellites to support this business, and the satellites are not compatible with each other. You can’t upgrade the existing base of hardware in the cars. How do you think about that satellite business? How many people work on the satellites?
It’s still the core of our business. We’re launching four more satellites over the next few years, and we have a constellation of satellites today that feed Sirius and XM, and actually, the satellites are designed now to address either network, so they’re interchangeable, which provides us with a lot of redundancy in orbit. There was just news today that cars are increasingly on the road for much longer periods of time. So we have an installed base of vehicles out there, 150 million or more, and many of them have only access to Sirius or only access to XM, which we refer to as low band and high band, respectively, but both of them get the same set of content today. That’s one of the things that we implemented shortly after the merger.
Wait, here’s my first feature request. My 2016 with only a Sirius radio, it can’t get Road Trip Radio because it’s in the high band.
I didn’t know that.
Right. If you look on Reddit, this is a common complaint about this one very popular station.
Very popular station, especially this time of year. I will take that back to the team and see what we can do.
All right. Results on Decoder. But this is the heart of the problem, right?
You’re heavily hardware-dependent in an increasingly software world, and I’m just wondering how you think about that particular kind of problem.
We have two avenues, I think, to address it. Increasingly, we are rolling out our 360L platform, and that allows for an unlimited amount of content. It’s rolling out slowly. About a third of our new car trial starts are 360L-capable today, but that’s only about 8 million cars on the road. So 8 of the 150 million. The other mechanism we have to address is, of course, streaming only. And we’re slowly building our streaming-only subscriber base. The vast majority of our subscribers today are focused on listening in the car with the radio that’s embedded in the car, but streaming allows us to leverage in cars that may not have a satellite radio or even in cars that do but maybe as a second listening point in a vehicle you just want to stream, so you can listen to the app through CarPlay, Android Auto, and increasingly, we’ll have better features there as well.
You just made some hires to improve that app and to improve the sort of online delivery of the content, right? Explain what’s going on there.
So last year, in January, we brought Joe Inzerillo on board as our head of product and technology. He came from Disney. He was very involved in the launch of Disney Plus, and as part of BAMTech, he was involved in a lot of other D2C video launches as well. He has built out a team at the company focused on our launch of our next-generation SiriusXM platform. It’ll start later this year with new streaming apps because, as we’ve talked about, the company started as satellite radio with no real digital connection to the consumer. So we’ve used a lot of traditional marketing methods to reach consumers, and going forward, we want to make sure that we can use all of the infrastructure that he’s building across commerce and identity, MarTech, and better data to provide better experiences for customers across all their touchpoints, whether it’s through marketing, in the product, of course, or customer service or anywhere else.
I want to come back to that because that sounds really interesting. If you just think about Sirius marketing today, you buy a car and you get mail until you give in and sign up, or the car is constantly showing you a preview. And you’re saying you’re going to build some other apparatus, which I want to talk about. But just at a really basic level, you have a new group that’s going to build a new online platform. They’re able to go and address a bunch of what are now essentially commodity services.
You hired somebody from BAMTech. They can go back to BAMTech and say, “We want to buy your platform and use it for streaming audio instead of video.” Or there’s a million other providers like AWS. On the satellite side, you described it as a proprietary platform, so there’s probably not the gigantic ecosystem of vendors and suppliers to say, “We’re going to build you a satellite radio service.” How is that cost split up in your head? As you think about running the company, do you have to employ a lot of proprietary satellite radio frequency engineers?
We do, and again, it’s a really core part of our business — making sure that we have engineers and experts on everything from the broadcast operations side through to satellite delivery, uplink, designing the satellites, working with Maxar as our third party that’s constructing the satellites, and SpaceX will launch them. But we’re really involved in the design of everything: the radios, the receivers, the chipsets, and working with third parties through that. And we have the same kind of challenges and opportunities everyone else has there: how do we make sure that we have multiple sources and we can continue to execute on that strategy? So we do have a lot of unique talent in the company to take advantage of that very proprietary pipe.
And one of the great things about satellite delivery or broadcast, if you will, is once the satellites are in the air, obviously we have the cost to maintain them, but it’s largely then a sunk cost. We don’t have any variable costs for delivery. And on the streaming side of the business, as we build that platform and it’s a supplement in the car for delivery for 360L, we will have more variable costs in the cost structure. But yes, there’s a set of common, at least underlying, principles in the architecture that we’ll be able to leverage as we build that.
So let me just decode that. You’re saying you spent a bunch of money up front on the satellite; you put it in the sky. That cost is cost, and it can service however many customers. And that’s fine.
And you go to the cloud, to broadband, and as you add customers, every customer is a cloud fee or a broadband fee or some other set of costs that come along with every new customer?
As you work on your projections and you say, “More and more, we’re going to go to broadband.” Is that something you’re factoring in?
Look, we have really strong margins on the SiriusXM side of the business. In part, that’s because of our licensing structure, but given our pricing and our high variable margins, it just provides us a lot of opportunity to be able to leverage different delivery mechanisms but still, on a marginal basis, ensure that our subscriber economics are very favorable.
Let me ask you the Decoder questions now because I think we’re right in the heart of it. How is the company structured? How have you organized SiriusXM?
We’re functionally structured, so my team of direct reports is made up of four individuals that I would say run the majority of the business side of the company. We talked about Joe Inzerillo, who is our head of product and technology. I have Scott Greenstein, who’s head of content. And then I have two people who are more focused on the commercial side of the organization: our chief commercial officer, Joe Verbrugge, who is responsible for engagement and onboarding of all subscribers or listeners across our businesses, and he has all subscription revenue as well; and then John Trimble, who runs all of our ad sales. They make up the business side of the organization, and then I have four individuals on my team who are responsible for more of the corporate functions. Barbara Daniel runs our strategy organization; and then, of course, our CFO Tom Barry; our head of people and culture, Faye Tylee; and then our general counsel, Pat Donnelly. And then I have a chief of staff who’s fantastic in helping me coordinate the team.
A functional structure here is kind of interesting because you do have these big groups that people can see — consumer-facing brands—– the radios and the cars but also Pandora and Stitcher and a bunch of famous podcast brands like Team Coco. Do you run them all centrally? Do you say, “Okay, here’s the Stitcher strategy, and here’s the car radio strategy, and that all rolls up to me, and I’m setting a strategy across them”? Because you would think that those would be more independent than not.
“At the end of the day, it’s about bringing content to listeners.”
The best thing is if we can have the functional leaders provide their expertise across these businesses. They’re not that diverse in the sense that we’re delivering audio content to listeners. At the end of the day, it’s about bringing content to listeners. I want our head of content, Scott Greenstein, to be focused on the best way to do that across all the platforms, and then our head of product and tech to be focused on the best way to deliver that content across all the platforms. I really believe in leveraging the expertise across these functional leaders, and then we work as a team, of course, to drive decisions that are going to improve the business across these different individual brands.
One thing I have learned over the past decade or so being a tech journalist is things that abstractly seem the same when you look at them on the internet are very different from a business perspective. When companies say, “We’re just going to use the internet to deliver audio the same way we delivered audio using FM or satellite,” my listeners are always like, “That’s just the same. What is the difference between a podcast and a radio show?” This is where I get to that thesis that the distribution actually has huge impacts in both directions, and the business of podcasting is way less lucrative than the business of FM radio. We just had Conal Byrne from iHeartMedia on the show, and he said FM radio is where the money is at, and he’s using the podcasts to help market that stuff. What is your most lucrative business? Is it still the satellite service?
Definitely. The SiriusXM side is the core part of the business, and I feel really good about where we are with that business because we have a highly profitable core business where our subscriber base is very loyal. We run an annual customer satisfaction survey. We’ve had the highest CSAT [customer satisfaction] score we’ve seen since we started running it over a decade ago, and that’s led to really low churn. Given that we have high variable margins, that flows through overall to higher EBITDA margins. And we are one of the most profitable media companies out there. That just gives us a lot of latitude to evolve the business going forward.
And I like our place in the advertising side of the business as well because Pandora really started digital audio advertising all these years ago, and we have a great, capable team there. A lot of ad tech with AdsWizz, which Pandora bought before we acquired Pandora, and we’ve been able to leverage that expertise, whether it’s the sales force or the technology, to have a bigger stake in the podcast business as well.
When you think about that structure and decision-making, do you ever consider moving a show from the podcast distribution network where the rates might be lower to the satellite distribution network where you have really high margins?
We think about that all the time. We have a lot of different models out there, and I’m not sure we’ve found the right one. There’s no consistent model across the board that we point to. Part of it ties back to what the creator or the talent wants. And there’s this balance. A lot of certainly talk or nonmusic content creators want broad-based distribution. It ties into relevance and audience growth, and we’re happy to support that. We have a lot of ad representation deals with talent like Ashley Flowers at Crime Junkie and her network at Audiochuck, where she wants broad distribution. So it makes a lot of sense.
Where we have changed the model over time is something like you said with Team Coco, where we actually bought the company from Conan O’Brien, and he’s stayed very close to producing his very popular podcast, Conan [O’Brien] Needs a Friend, which is broadly distributed, but we also have more exclusive content on SiriusXM. We have a linear radio channel that has elements from the podcast but also broader elements, and he’s created new content for that channel.
Over time, the objectives are to continue to support the advertising side of our business and grow that, and there’s certainly been a lot of tailwinds in podcasting there, but also find ways to serve the core SiriusXM business by bringing more exclusive content to that platform.
I want to get into that, especially the podcasting tailwinds because that seems to be just a convulsion that the entire audio industry went through, and maybe now we’re getting through the hangover. But first, more Decoder questions. You’re part of Liberty Media, a larger holding company that famously now owns Formula One. One Netflix show turned Liberty Media into the F1 company. Very smart. How does that work with Sirius? How do you integrate with that larger holding company?
[Liberty Media president and CEO] Greg Maffei is chairman of the SiriusXM board, and since they came on in 2009, we have had a really strong relationship with Liberty Media across the board. Really pleased to have them involved in the business.
One of the things that stands out about Sirius’ programming in general is the access to sports leagues. Every NFL game, every NBA game. Do you get preferential treatment from F1 because you’re part of the family?
We do a lot of work with F1 and the other businesses across the Liberty portfolio, but I wouldn’t say there’s any preferential treatment there. We just launched the Miami Studios, and we did some things together in Miami because it was right ahead of the F1 race there, and I think there’ll be more ways for us to collaborate. We work closely with Live Nation on a number of things. But no, I don’t think there’s any preferential treatment necessarily.
You have to figure out how to launch satellites and bring the cost down, invest, go hire people from BAMTech and Disney and build an app platform, argue across the family dinner table about F1 rights. How do you make decisions? What’s your framework?
I’m a big believer, because I have this team of functional experts, in bringing them together and hearing their individual opinions when we’re faced with a big decision. And I love the debate. We’ve always had a history, at Sirius, of kind of active debate on a topic. And then I’m hopeful that we’ll come to one path on our own. But obviously, if not, then I’m the decision-maker, and I really expect the team to jump on board with whatever decision we’ve made. I’m also a big fan of taking the contrarian view on something. So if we are looking at a decision, for instance, just to invest in this new platform, what happens if we don’t? And sometimes, the decision becomes pretty clear when you stack it up that way.
My personal favorite model is to just always have the worst ideas in the room. It’s surprisingly fun.
I doubt that.
Walk me through a moment where you had to be the decision-maker.
Probably in some of our content decisions because the challenge is decision rights at the end of the day. If we’re going to launch new content, I expect our head of content, Scott Greenstein, to come with that proposal and bring it forward and explain why it would make sense, but there are going to be opinions from others because the commercial team needs to execute on it and make sure that we’re getting the value from that new content. Or maybe, if it’s an advertising deal, we need the head of ad sales to be on board. But at the end of the day, those things may not be very clear across the team. So, in content decisions, I usually take a pretty active role.
One of the big themes with your content decisions over, I would say, the past year or so, is a really heavy emphasis on celebrity. You’re getting more famous names making podcasts. You acquired Stitcher and Earwolf. Those were a little more indie, a little more crunchy. There’s a pretty noticeable shift. You’re saying you’re in charge of content decisions. Walk me through that shift.
Howard Stern got the freedom he wanted to shape his show on Sirius
Some of it’s because the evolution of where we are with podcasting — that there’s a lot more content out there, and there’s a lot more attention around podcasting. It’s been a great funnel for new content for the business. If you go back to the early stages of our launch at Sirius, we brought Howard Stern onto the platform, and Howard had a massive platform through his core base in the New York City area, but then he was syndicated more broadly. It was time. He made the decision he wanted to come to Sirius, and that created a completely different trajectory for the business. It was really critical that Howard felt it was the right time for him and that he got the sort of freedom that he wanted to shape the show in the way that he wanted on Sirius.
That underlies a lot of the decisions we make. Developing the content pipeline that we’ve developed in podcasting has given us exposure to a lot more talent, and that goes back to the model that we talked about previously of finding talent that we can bring behind the paywall to bring even more value to our subscribers.
I did not expect a content moderation question in this interview, but you walked into it. With AM and FM, there’s the FCC. There’s a whole legal regime about spectrum scarcity, and that’s why the government can make rules. I don’t know how that plays with the satellite stuff, but I think it’s much less. And with the internet, all bets are off. The government can’t make any speech regulations. What are the restrictions in podcast content that you’re talking about?
I actually don’t think there are that many restrictions. I think it’s all about the business model and making sure that there’s enough brand-safe content that advertisers want to participate in. So I actually think there’s quite a lot of freedom in podcasting that looks more similar to what we have at SiriusXM. At SiriusXM, we want a broad set of voices. We are a platform for primarily opinions. We have the news networks on our service and political viewpoints across the spectrum, and we’re really proud of the diversity that we’ve been able to bring to our platform.
So you’re saying it’s the market that’s providing whatever restrictions on content? It’s what advertisers want to buy or what integrations, and that might be different on the satellite service than the digital distributed podcast service?
Back in the day, when Sirius signed Howard Stern, you just hired a famous guy, he brought a show over, he was expensive, but that’s just a talent deal. Lately, you’ve been buying companies. You spent $150 million on Conan’s podcast company. You spend a bunch of money on Crime Junkie, and you have an ad deal with Crooked Media that’s worth a bunch of money. You have 99% Invisible, Stitcher, Earwolf. You’re buying companies. Are you seeing the same return from buying those companies as from hiring a splashy celebrity to host one show?
“We bought Conan’s business, we bought Stitcher, we bought 99% Invisible ... I think the large assets have largely traded.”
Yes, we bought Conan’s business, we bought Stitcher, we bought 99% Invisible. The other deals you’re referencing are really ad representation deals, so we haven’t purchased the underlying IP. And where we’ve made the investment in companies and buying the underlying IP, we believe there’s real access to talent that makes sense for our platform and the economics work. On the ad representation side of the business, traditional podcast deals are basically ad share with a form of minimum guarantee, and we thought those were very important to our ad sales business because we are heavily invested in Pandora with the known operating platform, but we wanted to participate in the growth on the podcasting side and leverage all these capabilities that we talked about earlier.
Do you think those deals are working out? It feels like there was a real land rush to sign these kinds of deals. Disclosure: Vox Media has a podcasting network, they do it, too. Spotify spent a bunch of money, you spent a bunch of money, everyone else spent a bunch of money. Did it work out? It seems like that period has really cooled off.
If you look at our first quarter numbers, podcasting ad revenue is down about 2 percent, but podcasting was up 34 percent. So there is still a tailwind on ad sales in podcasting. I’m really pleased with the set of content that we have on the ad representation side. Regarding the deals, it was definitely very competitive over the last few years. I think the large assets have largely traded. And I think there’s an established set of distributors out there in the ad rep business who really love our content slate. We have five of the top 15 podcasts. We have a broad representation across comedy, true crime, and news, so we’re well positioned to take advantage of what I believe are continued tailwinds in podcasting — not only from a listener growth standpoint but from an advertiser standpoint because many brands are still sitting on the sidelines. It’s a small portion of their budget, but we really have seen, for instance, programmatic grow pretty nicely over the last several months, especially as advertisers’ budgets free up at the last minute. They can come in and buy across our podcast network, whether they want to buy within a specific show or they want to buy across an audience that would cover all of our podcasts.
You mentioned the AdsWizz acquisition earlier. You’ve got some technology where you can try to target listeners. This is the challenge in podcasting: figuring out how to get the right ads in front of the right people. And then you’ve got people like me who are cranky about host reads. Apologies to our podcast ads team. Those are very personal. You can’t target those. You can’t just self-service buy them. You’ve got to go find talent. They’ve got to actually read it. The platforms are standing right there to disintermediate you. You can just go buy the ads programmatically on Spotify, as opposed to coming to Sirius, or you can go buy the ads programmatically on YouTube, or go to the podcast host directly. How do you sit in the middle and say, “You should actually buy the ads from us in podcasting?”
Sure, you can find other programmatic solutions, but I think the direct sales component is really important still. If you look at somebody like Conan, who loves doing host-read ads, actually. Maybe you should spend some time listening to his.
There’s a real difference between being an entertainer and a journalist, but it’s all right.
They’re fantastic. He has a whole podcast episode on ads that he’s done, and he does it for the brands that he really believes in. We’ve had a lot of success with big brands on his podcast. I think it really depends on what the advertiser wants and if they want those custom integrations or they want exposure to live events. A lot of podcasters are doing more and more live events. We can bring that set of capabilities, in addition to programmatics. So it’s about a larger buy.
But is that more bespoke? “You’re going to spend more money, but you’re going to get Conan O’Brien reading the ads for you. You’re going to spend more money with us, and we’re going to make sure our hosts read across the network.” Or is it, “You’re going to spend more money with us, and we will efficiently reach more audience,” which is what the platforms would say?
Yes, and “you can reach more audience,” obviously beyond podcasts. There are podcasts distributed broadly across all other platforms besides ours, whether that’s Apple or Spotify or YouTube. And we’ll distribute that and represent those ads there. But also, for advertisers, we can provide you with solutions on broadcast, SiriusXM. It’s a relatively small portion of our ad revenue, but there are a lot of affluent customers listening on SiriusXM or Pandora, of course, which is really the largest piece of the puzzle today for us, where we can serve ads across the Pandora platform with a lot more first-party data and targeted ads.
So Pandora, to you, represents the direct competition with the platforms of Spotify and YouTube. You’ve got first-party data, listeners logged in, you know who they are, you know where they are, probably, and you can serve ads directly.
Is that growing? Are you investing in Pandora to make it grow into a true competitor to Spotify and the rest?
Pandora was sort of late to the game on interactive, so we have a nice number of subscribers, about 6 million. In terms of different tiers, we have radio service without ads, Pandora Plus, and then we have a small tier, which is fully interactive or on demand. But I believe the future for Pandora is really in the free tier. And we still have, if not the largest, one of the top two largest free music streaming services in the US. And we have a lot of strong ad technology and, again, the ad sales team selling across that platform. I think, from a consumer standpoint, we have an opportunity to continue to improve the listener experience, but in our sequencing of what we’re building... We can move Pandora to this new streaming platform, too, but that’s probably later next year because we have a lot of exciting things coming on the SiriusXM side between now and then.
Do you think the average consumer should know that SiriusXM and Pandora are the same company? Should they know about the connection there? Because, right now, they seem very different.
Not today, and I don’t think the average consumer does know that today. We’ve been able to bring the capabilities that Pandora offers to SiriusXM, which has been really important, whether it’s algorithmic programming or performance media or more digital CRM-type marketing capabilities that we just didn’t have at Sirius. Part of that is because Pandora’s had all this data for a long time, and at Sirius, we really haven’t had that much data on listening because, with the in-car, this satellite delivery mechanism, there’s no return path of data.
So I don’t know what you’re listening to in your 2016 car, but now, with 360L and, increasingly, our in-car customers streaming our service outside of the car and our streaming-only customers, we’ve got this really healthy dataset. So we’ve been able to leverage a lot of the capabilities from Pandora to help SiriusXM. And then, there’s been other ways that we’ve brought the brands together in different instances. We have the Pandora Now station on SiriusXM, we have SiriusXM content on occasion pop up on Pandora. So we’ve been able to bring content back and forth. But really, I think the biggest benefit has been on the capabilities.
I’m listening to 1st Wave, ’80s on 8, and Lithium. It’s a direct line. And then, every now and again, I listen to TikTok Radio, which makes me feel like a million years old.
That’s a great set of channels. You must have seven or so because that’s what the average listener has, seven channels.
Something like that. You have hundreds of channels, but you’re saying the average listener has seven. Pandora widens you out, right? You just start and off it goes.
Six million subscribers, ad supported. I listen to other streaming company CEOs, and they say, “Look, the future is probably some combination of subscription and ads together.” Netflix is a subscription. They’ve added an ads tier. They seem to be doing quite well with it. Disney’s headed that way. Is it the same in audio? Is it going to be the same combo where you pay a fee but you get some ads along the way?
Yes, I think so. I do believe, though, that SiriusXM, given the premium nature of what we’re providing, we won’t have a lot of ads in the service. We do have ads on our nonmusic content, but as we look forward to the flexibility that the new platform will provide us on, say, the commerce side, we can create a lot more packages. An ad-supported package might be compelling, especially to our core segments who maybe don’t want to pay $19 a month for our core service. So we have the opportunity to create more ad-supported channels, we can put ads in the music channels, and I think we’ll have an opportunity to test that when we launch what we’re calling our fast service later this year. It’ll start rolling out in 360L vehicles, and we can do more IP targeted ads through that as well. And that’ll give us a lot of flexibility to just test out different models with what content is available. Maybe it’s a free tier, maybe it’s just a lower-price tier, and adjust the ad load, for instance, or the content that’s available.
So we talked a lot about the sort of overheated podcast business, the big acquisitions, the flashy spending. You did have to have layoffs earlier this year, about 7 or 8 percent of your workforce. It seems like a lot of that was in the podcast part of the business. Walk us through what happened there and how you made that decision.
We came into this year knowing that both auto sales and ad sales were going to continue to be soft for some period of time. We’re a public company, we’re delivering value to our shareholders as well, and we’re very focused on making sure that we had discipline around our spending and we looked at every aspect of the cost structure. We did a lot of work last year to not have to address the people side, and we came into this year with a really thoughtful strategic approach to not only reduce costs but also to look at the business and make sure that we could be more effective and efficient. That meant some reorganization, not really at my leadership team level, but at teams below my leaders to bring things like analytics teams together on marketing or content teams together on the content side of the organization.
It definitely didn’t affect podcasting more than any other team. It was pretty evenly spread throughout the organization, and my main focus, having not really done one of these since the merger, was that we treated people fairly and respectfully and that we ended up in a place where the employees that are still with us feel very connected with what we’re doing and have a set of responsibilities that they’re excited about.
Do you think you would’ve had to cut less if you hadn’t splashed as heavily on the podcast acquisitions or the podcast spending?
No, I think the deals we’ve done in podcasting really set us up well for the future. In anything, you’re going to have to invest up front. I’m really excited about the monetization opportunities that are starting to play out and where we’re going to be in the future.
The thing that interests me about Pandora right now is that it’s in the middle of cutting-edge copyright lawsuits, and I’m an old copyright lawyer, so that’s captured my attention. A lot of comedians are suing Pandora saying that they’re owed two sets of royalty payments, both for the performance of their comedy and the underlying publishing rights to the written version of the comedy. There’s not a lot of precedent for that. You can see how, in the current marketplace of intellectual property, everyone is looking for the next payment because rates have just fallen so low with digital distribution. Where are you on that? Are you going to come to a resolution, or are you going to go all the way to a trial?
I can’t comment on the status of the litigation, but I will tell you Pandora’s been a great platform, supportive of comedians for many, many years. And with the comedians whose works we licensed from the labels, in many cases, we always thought we’ve had the whole right, whether it’s recording or the writing of the underlying jokes. As this litigation started, we did see some comedians leave the group that are participating alongside the plaintiff because we had to take down their works until it’s resolved. So we have had some comedians break away from that in order to get their content back up on our platform. But I really believe in comedy; it’s been one of the biggest drivers of listening outside of music on Pandora and is certainly core to our SiriusXM service as well. So we will continue to work through it and hopefully come to a resolution.
You run one of the most interesting content businesses out there because you are so regulated in terms of rights and payments that you send out to people. Satellite radio licensing fees are written into federal law. We’re kind of staring down the abyss of copyright law with generative AI and being able to fake voices and names and likenesses and fake Drake. Are you just waiting for the bomb to go off? Are you actively participating in conversations about what the future rate structures should be?
Our core music licenses on the SiriusXM side of the business are licensed through the US Copyright Royalty Board. We have an established rate in place through… I believe it’s the end of 2027, and we’re paying on the musical works that we play, and we aren’t really involved in the generation of music. So while I think AI will be very positive in terms of its impact on our business, there are not really the same issues that we might see in other platforms.
You have TikTok Radio. AI Drake explodes on TikTok. Would TikTok Radio play the fake Drake song?
I don’t know. I guess we haven’t really crossed that bridge yet, but we’ve always been alongside the artists on this, and I think we’ve had a really strong position in breaking new artists and building their careers and do want to continue to make sure that we’re doing that.
The TikTok Radio piece of the puzzle is actually really fascinating to me because there’s so much content on TikTok where it’s just a creator or someone like Charlie Puth just screwing around, and there’s no associated label to go negotiate with. There’s no infrastructure of licensing that content and making sure payments actually reach the creators. But a lot of it actually does end up on TikTok Radio. Do you have a team that’s just like, “We’re going to go figure out how to license this stuff?”
I mean, TikTok Radio is really a curated experience. So we are highlighting the content that has become popular on TikTok, so we have the opportunity to decide what is on the channel and what’s not, right?
But I hear remixes on there that, if I were a programmer, I could not tell you even where to begin to go get the rights to a popular song that’s had a remix and has now been sped up by a third person. That just seems like a puzzle of rights management that somehow ends up on the radio.
All of our content is licensed either through the US CRB on the recording side or with direct deals we have with the publishers, mostly through the collectives, right? ASCAP, BMI, etc. So those musical works would be licensed through those deals that we have in place already.
Let’s talk about cars. I’ve been very good at not talking about cars this entire time. That’s all I want to do. I think you have the most unique perspective on the car industry. The way I think about it, you’ve been making software for cars as the only third-party app vendor for two decades. You buy a Ford in 2004. There’s no app platform in there, but Sirius’ code and hardware is in there. We’re undergoing a massive shift in the car business. Sirius’ business is really exposed to the car business. You’ve mentioned it several times. The first thing I noticed is that the industry is selling fewer, more expensive cars. The trend all the car makers seem to be on is instead of a lot of $50,000 pickup trucks, they’re going to sell fewer $100,000 pickup trucks. That’s where they’re going. That just reduces the number of cars you can activate Sirius in. Is that something you’re worried about?
“Every time somebody buys a new car, it’s another opportunity for us to get a new subscriber”
The car market is certainly our biggest funnel. We are in just over 80 percent of new car sales and just over 50 percent of used car sales. So every time somebody buys a new car, it’s another opportunity for us to get a new subscriber, or we have existing subscribers trading their cars or adding a car to the garage. It is really impactful on our business, and we’ve seen subscriber additions slow as a result of that because of the last few years. We’re starting to see supply issues alleviate, some of the supply has grown, particularly for the domestic brands. That’s going to continue. And in the automotive industry, there have been these cycles in the past where everyone likes to say, “Well, the margins are really high. Maybe the industry will pull back on inventory levels.” But at the end of the day, it’s pretty competitive, and they want a share. So I expect that, over time, as long as the demand is there and supply starts to come back, sales will ultimately rise because the automakers want to go after that incremental sale.
Car company CEOs love to appear on Decoder. I love having them. There’s a huge shift in how we perceive of cars. They’re becoming much more integrated computing devices, especially with the shift to EVs. One of the reasons they’re excited about that is because they are looking at Apple’s margins in the App Store and saying, “I want 30 percent of every transaction that happens on this screen, too.” You’re right in the center of that, right? You’re saying, “Put my hardware in your cars. Here’s a subscription fee to a very premium service.” Up until now, it doesn’t seem like there’s been a big revenue share going on there, but I can imagine every car maker is going to say, “Why would I work with you when I can go sign a new kind of deal with Spotify and get a percent of their cut?”
There is a lot of revenue. From day one, we’ve shared in the subscription revenue with our partners, and I think not only does that differentiate between us and other media services because we have the margins to be able to support that, but it also helps us engage with the OEMs to make sure that we have the same incentives, that the service is prominently displayed on the screen, that we’re getting information on the customers so we can market, sending that direct mail that you get, or emailing customers, making sure that the trial’s actually on, and it’s very easy to get into the service after you’ve bought the car. You have a lot of other things on your mind when you buy a new car, so we want to make sure the service is on and working when you step into that new vehicle and you get to experience all the content. So the revenue share has been key to that in terms of working closely with the OEMs on what the future is for SiriusXM.
Do you think that revenue share is going to provide a moat against competition? You said you had high margins because you’re the only premium offering in a car. There literally wasn’t another one. Now, you’ve got people like me with their phones on their dashboard running Apple Music. You’ve got CarPlay and Android Auto that exposes that directly in the interface. Is competition from phones bringing down your margins in the car?
No, it’s not bringing down the margin. Ultimately, we’re agnostic about what distribution we use. So we will always have an integrated radio experience in the car with the receiver, and ideally going forward, our best platform is 360L in terms of providing the best experience for the consumer because satellite delivery allows for uninterrupted delivery of our service across the US. Of course, with modems, there are always going to be dropouts in less heavily networked areas of the country, whether urban or less dense areas. So that is a real benefit to satellite delivery. But with the IP delivery, we can actually customize the service in ways that all the other media services can do.
So yeah, the infotainment systems in the car have been evolving for years, and we’ve really been at the forefront of that, making sure that we’re designing for those new infotainment systems, working closely with the OEMs but also, in some cases, working closely with, say, Google, who’s launching Android Automotive, the operating system in the cars. That’s going to give us a lot more flexibility, too, because we can update our service quarterly or so just like every other media app, but also we’ll be represented in multiple places in that screen as well.
So, this is one of my favorite ways of thinking about cost. You’ve got two buttons on a screen, and you push one button, call it Nilay’s Music Service so you’re not talking about your competitors directly. But Nilay’s Music Service is an IP-based streaming service. You might say it looks a lot like Spotify. And I push that button, and I’m just trading on all of the mobile broadband infrastructure that exists. And I might even be trading on a consumer’s own broadband subscription so I don’t even have to worry about that, and that costs 10 bucks a month.
I push the Sirius icon on that same screen, another interface opens, and now, I’ve got your satellites in the air, I’ve got custom hardware in the car, I’ve got a 360L system in the background running. That’s all just more expensive.
But all that’s happening to me as the consumer is I’m clicking a button and listening to music. How are you going to compete against the fully digital platforms that are trading on other people’s connectivity?
So more expensive in the sense that we’ve launched satellites and we’ve got this infrastructure? Because that’s already in place.
Well, just the cost of SiriusXM is more than the cost of Spotify.
Oh, to the consumer.
Yeah. So as we develop this new... our next-generation SiriusXM platform, we’re trying to solve for a few different key pain points, which is going to allow us to broaden the universe of our prospective subscribers. And those are price, control, and discovery. That’s what we hear time and time again.
So the experience itself is going to make sure, because it’s so critical, whether you’re listening to our service through the mobile app, through CarPlay or Android Auto, or you’re leveraging an embedded streaming version, or an embedded satellite plus streaming and 360L that, ultimately, we need to provide more control and discovery to the consumer.
Because, in the past, it had been all about broadcast and just turn the dial to find new channels, and that’s not what many younger consumers certainly want to do.
So, providing these more enhanced recommendations is going to be key to that. We have so much great content, but it’s very hard to get consumers into content if they don’t know what we have.
And then, providing links from that embedded radio experience and the car out to the streaming apps so that discovery can happen there, which may be a better place to facilitate it.
And then those behaviors, those listening behaviors, can come back into the car. As consumers experience different channels or different content outside of the car will bring those experiences and behaviors back into the car.
Ford is going to go from here to there with their software stack. They’re supposed to be moving to Android. Who knows? They keep launching cars without it. I need to get Jim Farley back on the show and ask him about it. GM just redesigned their interface to say, “We don’t want CarPlay and Android Auto anymore, and our new EVs going forward won’t even have it.” Tesla exists. My new Jeep that has 360L runs a forked version of Android that is based on Alexa and TomTom maps for some absolutely bonkers reason. That’s a lot of software ground to cover. That’s a lot of platforms to support.
One extremely universal theme on the show is that once you start investing in software, the costs just escalate and never come back down. How are you thinking about that — supporting 95 different car makers and their bizarro platform ideas?
That’s where something like AAOS [Android Automotive operating system] comes in, which I believe about 70 percent of OEMs have said that they are going to adopt. We’ll see. That helps us streamline the development process because it’s basically one app, and we can distribute it across all those players.
But that actually hasn’t come true yet. Ford announced that they were moving to Android almost 18 months ago, and the new Mustang is out, and it still runs QNX with Unreal on top of it. They’re still looking at this and saying, “The big platform companies are sitting on top of us.”
Last year, almost exactly a year ago now at Apple WWDC, they announced a version of CarPlay that would take over all the screens in your car.
Do you see that happening? Do you hear that rumble in the industry — that the car makers once thought they would give up the car interface to Apple and Google, and they are actually realizing that it’s not the right choice?
I think the automakers don’t want to give up control, so they’re trying to find the areas where the Big Tech companies can assist them, but ultimately, they can still control the user interface and try to participate more in the services business.
So we have to be really flexible, and we really have been for 20 years, because we’ve had to design for these different automotive UIs since inception, and it was more hardware and less software, and over time, it’s shifted to even more software.
We’ve already had the infrastructure, the team, the resources in place to be able to design across multiple types of implementation. I’d say we probably had less control over what that looked like than we will going forward as long as the OEMs tend to coalesce around a few different solutions, but we’re going to have a lot more flexibility no matter what, especially as we design for these new solutions that are coming forward.
Here’s an extremely loaded question. Do you use CarPlay in your car?
I do, yeah, but I move back and forth between the SiriusXM radio, and then I’m using CarPlay for my navigation.
I don’t even use CarPlay in my car. I just mount my phone on the dashboard, which my Vergecast listeners know. They think it’s the dumbest thing in the world, and I should just use CarPlay.
But that shift where I would rather have two screens with your interface on one screen and a map on the other — no car does this well. Are you happy with how your software is being expressed in the car, and do you think the car makers are going to give you the tools to get to where you should be?
In part, it comes back to two things: revenue share, which certainly motivates, and consumer demand. So one of the biggest complaints we hear quite often from consumers who have a number of these startup EVs — high end, new introductions to the market on the EV side where we aren’t present today, though we just launched a beta version in Lucid — is that SiriusXM isn’t there, and it’s not easy to use necessarily.
“We have a lot of affluent customers who just come into their new vehicle purchases and expect us to exist there”
I think that works in our favor, especially because, again, we have a lot of affluent customers who just come into their new vehicle purchases and expect us to exist there. So I do believe that’s going to be a force to help us navigate this with the OEMs to ensure that our presence on the screen is very obvious and that customers can get right into the service easily.
You mentioned you didn’t have a lot of control over your user interface before. That makes sense for a million reasons. Now, we’re just putting full-on tablets in the middle of cars, for better or worse. Those are app platforms. You can design the app however you want. Is your vision that the app on a phone, the app on a car, the app on an iPad should all look and work the same?
There should certainly be continuity, but we’ve designed for these in-car experiences for a long time, and we understand the security and safety guidelines through NHTSA for in-car implementation, so I think it’ll naturally be different in the car. We can do a lot more in the apps in terms of allowing for control and discovery, but we want the continuity across the platform. What we’re building now will allow for that, whether you’re listening through the apps or in the car or in some ways through connected devices in the home as well.
Where do you think the endpoint of this big shift to cars as computers is? Is it, “Finally, we’re going to get rid of the steering wheel and self-driving happens and everyone just is in a world of screens happily shopping in their cars,” or is there a different kind of endpoint?
We’re certainly talking a ways out into the future, right? It’s really interesting because, I’d say, 10 years ago, everyone was talking about autonomous, and it was the biggest topic for the automotive industry, and you don’t really hear as much about it anymore. It’s all about electric, and we’ve got our hands full, I think, as a country and an industry and just rolling out electric vehicles.
I’m biased. I’m at an audio company, but I do believe that intimate listening experience to audio in a car, something that we do phenomenally well, people are just passionate about it — whether it’s 1st Wave or Richard Blade or you’re passionate about the host and the curation that exists on our service, that audio listening experience, even if the steering wheel goes away, is going to continue for a long, long time.
This is the threat, though. I was asking about it nicely, and you were talking about it directly, which is: you take out the steering wheel and now everyone is watching YouTube on their phones. Do you see that as an existential threat? Do you have to market against that?
Certainly not today, and I would say it’s going to be a lot more like two decades into the future, and even then, it’s a small amount of cars coming out, and I do think, at some level, it flies in the face of what consumers really want. There are a lot of consumers out there that want the driving experience, and again, it’s been all of these years, we’ve all been trained as consumers on audio in the car, and it’s a very special connected experience to be listening to, whether it’s your music, your favorite channel, or your favorite talk host.
It could be Howard or Andy Cohen or Megyn Kelly. We have such great talent on our platform that I think those connections and that habit is very well established. It’s going to take a lot to disrupt that.
Where do you think the combination of podcasting and broadcast comes together? Because what we’ve talked about since the start of this conversation is: at the end of the day, the consumer kind of doesn’t understand the difference between the distribution, but your business definitely does, right? Your business knows, “With the satellites in the cars, there’s an entire well-built, lucrative business model with high margins.” And then, over here on podcasting, we’re still inventing ad tracking tech and marketing technology, and that’s a wide-open space.
If self-driving is two decades away, you’re not worried about that — the convergence of those things seems like tomorrow, right? We’re actually asking people to pay a lot of money for a hardware streaming system when the cars have Level 2 driving, and you can maybe screw with your phone a little bit more. That’s going to get more and more squeeze. Are you thinking about that?
We’re thinking about it in how we’re designing this platform because it comes back to, again, addressing these gaps we have around control and discovery, but fundamentally, we are a curated audio experience. I think what our customers love most about us is that we do help guide you through the content you’ll love, and we have a set of music channels that our consumers love. We have a set of talk audio content that our subscribers love. That may be in the form of podcasts or just on demand. It could be live shows that are adapted to on-demand and show up as a podcast.
So I think as long as we are providing that guided experience and offering consumers a more curated experience so they can help find the content that they love, that’s where we provide a lot of added value versus a lot of other platforms, where you’re just doing fundamentally a lot more work to find things that you love.
All right, Jennifer, this has been great. Thank you so much for giving me all the time. Thank you for indulging the car talk. Truly, I could have done the full hour on just what’s going to happen to the car.
We can come back and do that.
You said there’s a new app coming. Give us the timeline. What’s next for Sirius?
We’re really excited about this next-generation experience. It starts with a launch later this year in the fourth quarter with a set of new streaming apps, but also the first step in re-architecting everything, all the foundational capabilities — really addressing every consumer touchpoint with better commerce, identity, MarTech data capabilities. That foundational experience will improve relevance, ease of use, and value for our subscribers and prospective trialers and really go toward addressing these pain points that we’ve talked about, whether it’s flexibility on price or more control and discovery. And it’s going to help enhance the streaming experience — if you’re a streaming-only subscriber or if you’re an in-car subscriber, whether you’re leveraging satellite only or 360L, it’ll help with your listening behaviors outside of the car, which are delivered by streaming or the streaming infrastructure that supports 360L. And just, again, enhancing value for our subscribers by providing easier navigation of our service and more discovery of all the great content we have.
I think it positions us really well for the future. We have a really strong foundation financially but also a very loyal, satisfied set of subscribers that we can build off of as we launch this next experience.
Amazing. Looking forward to that. When the app launches, you’re going to have to come back and tell us all about it.
Can’t wait. Thank you, Nilay.
Thanks so much.
Decoder with Nilay Patel /
A podcast about big ideas and other problems.