Netflix, iHeart, and the Metrics That Actually Matter

Reid Mercer: Netflix and iHeartMedia just handed us a case study. Martha Stewart, Kate and Oliver Hudson, Lele Pons all locked into exclusive video podcast deals. Reuters and Variety both covered the announcement today. But the structure of the deal is where the real story lives. We'll get into what iHeart is actually taking on versus what Netflix walks away with. We want to hear from you. Submit questions via the web form in the description or give us a call at 747-234-2678 and leave your question. Don't be shy. Our AI assistant makes it super easy. Netflix and iHeartMedia just expanded their exclusive video podcast partnership, iHeart's press release announced this morning. Variety confirmed the talent slate: Martha Stewart, Kate Hudson and Oliver Hudson together, and Lele Pons. Netflix gets the exclusive video window on all of them. Read the headline and it looks like Netflix is going bullish on podcasting. Look at the structure and a different picture emerges. Netflix isn't producing these shows iHeart is. iHeart handles the talent relationships, the production infrastructure, the audio distribution. Netflix takes the video window, and crucially, the audience data that comes with it. That's the part the press release buried in paragraph four. Think about what Netflix actually gets you that it couldn't build itself. It doesn't need iHeart's production capability. Netflix is spending years building original content pipeline. pipelines. What it needs is a shortcut to an engaged podcast audience without paying for development risk. iHeart absorbs that risk. Netflix shows up for distribution. I've watched this movie before. Cable operators in the 1990s didn't produce ESPN or CNN, they just controlled the window into your living room. The network did the work. The distributor owned the relationship with the viewer. Netflix is running the same play. The content flows up, the data stays downstream. Now, is the talent slate actually the right bet? Martha Stewart, Kate Hudson, and Oliver Hudson, these are household names, that's real, but celebrity-driven video podcasts have a mixed track record. The question isn't whether those names drive initial clicks (they will), the question is whether they drive subscriber retention, which is the only metric Netflix actually cares about. Follow the money on the exclusivity structure. iHeart still distributes the audio version everywhere. Netflix only owns the video window, so Netflix is betting that video first podcast consumption is where the growth is and that having Martha Stewart or Kate Hudson on your platform in video format moves the needle on subscriber value. That's a speculative thesis. Could be right. Could be that Netflix is paying for marketing rather than content. Reuters noted this morning that this is an expansion of an existing partnership, not a new relationship. The first phase apparently worked well enough for both sides to go deeper. That's actually the most telling detail: iHeart gets a premium distribution window that makes their talent deals more attractive. Netflix gets video content at a fraction of what original programming costs. Both sides have incentive to call this a success, regardless of what Speaker 2: the numbers look like. Reid Mercer: So what the numbers say. And that's worth flagging. When a deal is structured so that both parties are incentivized to report it as working, you should stress test the metrics before you accept the narrative. What does engagement actually look like on Netflix-hosted video podcasts? We don't have those numbers publicly. Netflix doesn't share them. The strategic read: Netflix is establishing a beachhead in video podcasting through celebrity celebrity talent because it's cheaper than building from scratch and faster than waiting for organic discovery. iHeart gets distribution credibility and talent leverage. The exclusivity window is the real asset being exchanged here, not the content itself. Which raises a question worth sitting with if celebrity names are doing the heavy lifting in these deals, driving the contracts, justifying the exclusivity windows. anchoring the network pitch. What happens when you try to build a whole network on that logic? Does celebrity actually retain audiences or does it just win the initial signing? So the Netflix-iHeart deal uses celebrity names to anchor the pitch: Adam Ray, Martha Stewart, Kate Hudson, talent as the headline. But there are two very different bets being made out here, and conflating them is how you end up with a bad content portfolio. Take Adam Ray. Variety reported the show premieres on Netflix June nineteenth. Comedy interviews, character work, a comedian who's already built an audience. That's a content bet. You're buying a proven voice and putting it on a bigger platform. Then look at Alex Cooper's Unwell Network (Variety also covered this Thursday), Unwell just signed Georgia Hassarati and Isabel Timerman's show Girls Disrupted. Cooper's assembling a Gen Z-focused network around talent that already has audiences locked in. That's a distribution bet wrapped in a talent announcement. These look the same from the outside. Both involve celebrity names. Both generate press, but one is acquiring reach and the other is acquiring a content asset. The unit economics are completely different. I've seen this pattern before goes back to cable bundling in the 90s networks would sign talent with massive upside and bury the downside in the production line the talent got the headlines the network got the liability if the show didn't perform ask yourself who holds the risk if Adam Ray's audience doesn't migrate from wherever he built it to Netflix's podcast tab Celebrity name winning a deal isn't the same as a celebrity name retaining an audience on a new platform. That's the question nobody's asking at the press release stage. And Cooper's Unwell play is smarter structurally, at least on paper. She's not betting on individual shows; she's building network economics, multiple talent relationships, shared audience infrastructure, a brand that attracts Gen Z creators before they get big enough to get poached. The moat is the network, not the name. The celebrities aren't the product; the audience aggregation is. is the product, the talent is the acquisition cost. Which brings up a question that gets more urgent the more money flows into these deals: how do you actually measure whether the bet paid off? Because right now, content investment is scaling faster than the measurement infrastructure can handle. Spotify just moved to redefine what a play even means, and that one definitional shift touches every performance metric in the space: ad buyers, creators, network executives. If the measurement standard changes, so does the scorecard on every single one of these deals. The measurement question is exactly where this gets real for ad buyers. Spotify just redefined what counts as a play, and the buried detail is what it means for every CPM deal negotiated against the old standard. Spotify's announcement this week moved the threshold to 30 seconds. Watch or listen for 30 seconds, that registers as a play. Spotify's own audience behavior data shows 30 seconds is a reliable indicator of... of active engagement. Clean story, right? Here's what the press release didn't say: Every ad buyer who locked in CPM rates against historical play counts is now looking at a different denominator. The number of qualifying plays goes down, effective CPMs go up, unless the contracts explicitly defined play by the old standard. Did yours? And Podnews flagged the tension immediately-the industry doesn't have consensus here. Sixty seconds is the rival threshold that some analytics providers still use. This looks like a standards debate. It's actually a power play-whoever's definition of a play becomes the default, controls how performance gets reported across the market. Spotify has scale, they can move the needle on this. Now, flip that against something counter intuitive: the Radio and Television Business Report covered new research this week showing that more than half of podcast listeners who routinely skip ads have still acted on one, visited a website, searched a product, made a purchase. YouGov put the number at sixty percent of podcast consumers taking action after ad exposure, even with habitual skippers in the mix. So we're simultaneously debating whether thirty second seconds of listening is meaningful engagement, while the data shows incomplete exposure still drives brand outcomes. That's a strange place for the industry to be sitting. What it tells sophisticated ad buyers is this: completion isn't the only metric that matters; awareness generated during a partial listen is still awareness. The measurement conversation needs to separate reach from completion from action, and most current frameworks collapse all three into a single play count. The deeper issue is who sets the standard. In the 90s Nielsen didn't just measure television ratings; they defined what counted as a viewer. Whoever held that definition held leverage over every upfront negotiation. Spotify is making a similar move here. If their 30-second threshold becomes the industry default, they've effectively set the rules for how audio inventory gets priced. That is not a product update. That is a market structure move. Ad buyers with volume commitments should be pulling contracts right now, not because the numbers collapsed (the YouGov research suggests brand impact holds even with skippers), but because the goalposts shifted and most deals weren't written with that flexibility. And then there's the cost side of the equation. Because all of this measurement infrastructure assumes the production cost baseline stays roughly constant, it won't. Rebel Audio just went public beta with an AI production platform, and Geico deployed an AI-generated Gecko as an actual podcast guest. The economics of who can afford to play in this space are about to change. AI and production cost, That's the variable that changes the math on everything we've talked about today. That's the episode, four stories, one through line. Every deal, every metric, every AI stunt has a structural logic underneath it, and the headline almost never shows you what it is.