SiriusXM's Cash Flow Problem Is a Subscriber Problem — But Maybe Not Anymore

SiriusXM's CFO is due at a Wall Street conference in New York on September 9th. The press release announcing the appearance is two paragraphs long and entirely generic. The news behind it is far from it.

After four years of falling revenue and three consecutive years of self-pay subscriber losses, SiriusXM's underlying business has shifted in ways its stock price does not yet reflect. The satellite-radio monopoly that once defined the company is shrinking. The audio-advertising platform it has been building in parallel is growing into something worth examining on its own terms.

The numbers that get most attention tell only part of the story. SiriusXM's revenue has declined every year since 2023, from nearly $9bn to the $8.53bn the company now guides to for full-year 2026. Its self-pay subscribers — the ones who actually write a cheque, rather than riding out a free trial — fell by 445,000 in 2023, 296,000 in 2024, and 301,000 in 2025. The total subscriber base, padded with free promotional plans from automakers, has hovered at roughly 33 million, masking what is happening to the paying core.

Yet the rate of decline is slowing. In the first quarter of 2026, self-pay losses narrowed sharply to 111,000, compared with 303,000 in the same quarter of 2025. By the second quarter, the company posted its first positive self-pay net additions in four years: a gain of 22,000. Self-pay monthly churn fell to 1.4%, a record low. Those figures do not signal a renaissance. They suggest the bleeds are being plugged.

The mechanisms behind the improvement are worth dissecting. A large part of the Q2 gain came from companion subscriptions — the add-on plans that let SiriusXM customers listen outside their cars on phones and tablets — which contributed 123,000 incremental additions. Extended-duration dealer plans and continuous-service initiatives also helped. In other words, the company is keeping existing customers paying for longer and encouraging them to subscribe to more services. It is not attracting a new wave of buyers. The stabilisation is real but narrow.

The more consequential shift is happening on the advertising side, where SiriusXM is quietly assembling what may become the largest ad-supported audio ecosystem in America.

The subscription business, the one investors fixate on, generated about $6.42bn in 2025 — down from nearly $6.84bn in 2023. But the company's advertising revenue, spread across Pandora, its podcast network, and the SiriusXM platform itself, held remarkably steady at roughly $1.77bn each year from 2023 through 2025. While the satellite subscription side has been bleeding, the ad business has been flat rather than falling. In a company whose total revenue is shrinking, that is not nothing.

Podcast advertising is where the growth actually is. Podcast ad revenue rose by 12% in 2024 and then by 41% in 2025, after SiriusXM invested heavily in exclusive content deals with names such as SmartLess, Call Her Daddy, and Mel Robbins. The company describes itself as the number-one podcast network in the United States by weekly reach. That is a self-description, but the revenue numbers are not fabricated.

The largest single step yet came in April 2026, when Google announced that SiriusXM Media would become the exclusive U.S. advertising representative for YouTube's audio advertising inventory — the ads that run when someone is listening to podcasts, talk shows, and music on YouTube rather than watching them. A joint study with Edison Research found that over 212 million monthly listeners in the United States engage in audio-first content on YouTube. The deal, which launches in the autumn, gives SiriusXM the right to sell guaranteed ad impressions on that audience at scale.

The strategic logic is clear. Audio accounts for roughly 30% of media time but attracts only roughly 4% of ad spend. SiriusXM is attempting to redirect budget into the gap. The YouTube deal expands its combined portfolio to 255 million monthly listeners — nearly 90% of the U.S. population aged 13 and older — and turns SiriusXM Media into the single point of purchase for ad-supported audio across YouTube, Pandora, SiriusXM, and its podcast networks. The company is not merely selling ads for its own platforms anymore. It is building a marketplace.

Which brings the analysis back to the stock. SiriusXM trades around $24, giving the company a market capitalisation of approximately $9bn. It guides to roughly $1.375bn in free cash flow for 2026 and $1.5bn by 2027. On those numbers, the free-cash-flow yield sits near 15% — an enormous figure by any standard. The quarterly dividend of $0.27 per share, or about $1.08 annually, yields roughly 4.5%. The company has been returning capital steadily and has brought its net-debt-to-adjusted-EBITDA ratio down to 3.4x, within its targeted range.

The market is not celebrating this. Shares fell 5.8% after the Q2 results and the guidance raise, and are down roughly 12% over the past year. The reason is obvious enough: investors are pricing a declining subscriber base, the long-term threat of connected-car platforms favouring Spotify and Apple Music, and genuine uncertainty about whether the advertising business can grow fast enough to offset the subscription side's structural erosion. These are not imaginary risks.

The trouble is that the market's model of SiriusXM may still be the old one — a satellite-radio monopoly whose fate is tied to whether cars come with a SiriusXM receiver. That model is not wrong. It is incomplete. The company that generates over $6bn in subscriber revenue from roughly 31 million paying customers is also the company that generates $1.77bn in ad revenue from 255 million monthly listeners, with podcast advertising growing at 41% a year and a new exclusive pipeline into YouTube's audio inventory launching later this year.

The question for the investor is not whether SiriusXM's satellite subscription business is declining. It is. The question is whether the advertising platform it has assembled — and is still assembling — is worth more than the discount the market applies to a fading radio company. A free-cash-flow yield of 15% implies the market expects cash flow to fall, or never to compound, or both. That is a view, not a fact. The YouTube deal alone, if it meaningfully grows the ad business, would change the revenue composition even if subscriber numbers never recover.

There are reasons to be sceptical. The ad industry is cyclical, and podcast advertising has grown from a small base. The YouTube deal has not been priced or quantified in any public filing, and it faces execution risk in an ad-tech landscape where Spotify already operates as a walled garden and Google itself could reclaim the sales function. Self-pay subscriber growth of 22,000 in a single quarter, driven largely by companion plans, is not a trajectory. And the connected-car threat — streaming services integrated directly into vehicle infotainment systems, making SiriusXM's satellite signal redundant — is structural and irreversible.

Yet a 15% free-cash-flow yield is not the kind of number that persists in a functioning market unless the eventual outcome is better than the price implies. SiriusXM's satellite business will keep shrinking. Its advertising business may or may not grow enough to replace the gap. The evidence available today — record-low churn, stabilising self-pay metrics, a rapidly growing podcast ad franchise, and an exclusive YouTube deal — points in one direction. The stock price assumes the other.

The market will decide which evidence wins. Until it does, the discrepancy between the cash flow and the share price is the investment to think about.