Podfade Killed Three of Every Four New Podcasts

Your new favorite podcast is probably already dead. That is not a guess. Podchaser tracked every podcast whose first episode launched in 2025, all three hundred thirty-three thousand seven hundred twenty-three of them, and the company reported that roughly three in four had gone quiet by October. The study puts a hard number on podfade, the industry's word for a show that stops publishing new episodes without ever formally ending, and Podchaser documented the scale of it in an analysis published this week.

Podfade almost always arrives early, and it arrives quietly. More than a quarter of last year's class published exactly one episode and never came back. Nearly half of the shows that went silent published nothing after launch day. The median quitter managed three episodes and eight days before the feed froze. The typical version of podfade is a shrug.

Podcasting guides love to warn that most shows die before episode seven, and the data says that line has never meant much. Stopping at exactly seven episodes was about as common as stopping at six or eight. It is true that most failed shows stop before episode seven, but only because so many of them stop at episode one.

A lot of the wreckage, it turns out, was never really a show. Someone tests a hosting platform with a voice memo and never returns. A local business publishes a feed that reads like an ad. Podchaser found tens of thousands of feeds with a hosting platform's default-style name, plus a few thousand automated batch feeds, mostly public-domain audiobooks read chapter by chapter and cloned language courses, where more than nine in ten are still technically publishing. That leaves podfade looking less like a mass extinction and more like a failure to launch.

Strip those out and the picture brightens. The study defined a serious launch as a show with at least three episodes released over more than one day, a real name, and no batch automation, and about one hundred ninety-three thousand shows qualified. Among those, about a third were still publishing, and nearly three in five were still going six months after launch.

How the survivors outrun podfade

Shows that book guests last longer, and the pattern holds in every version of podfade Podchaser sliced. Among serious English-language launches with twenty-six to fifty-two episodes, roughly two in three shows that book guests were still publishing, compared with half of the shows without guests. The likely reason is simple: every guest brings a new conversation, a built-in audience, and a reason to share the episode.

Longer episodes go with longer lives. Only about one in five serious launches averaging under ten minutes an episode was still publishing, against more than two in five averaging forty to sixty minutes and nearly half averaging over ninety minutes. The biggest jump sits between under ten minutes and twenty minutes or more. Short shows are cheaper to make, but they are also where most of the quick experiments live.

A steady, unhurried schedule beats a daily grind. About forty-four percent of serious launches publishing roughly every two weeks were still going, against about twenty-eight percent of shows that published daily or close to it. The daily pace is hard to keep up, and the burnout shows in the data. The Netflix Barstool podcast deal pulled big sports shows off YouTube this month, which is the opposite problem: established shows with real audiences are consolidating while the long tail goes silent.

Why the surviving shows are worth more than ever

The remaining quarter of the class carries real influence. Acast's Podcast Pulse 2026 study surveyed four thousand five hundred podcast fans across thirteen markets and found that more than two in five trust brand recommendations from their favorite podcaster, up nine points in a year. Trust in journalists sat at about three in ten, with YouTubers and celebrities trailing behind, according to a report on the research published this week. Mike Koenigs, a founder who commented on the research, urged founders to become the public voice of their companies, saying trust has moved from the brand to the person people listen to on their commute. Nearly nine in ten fans follow or interact with their favorite host outside the podcast, which is why Joe Rogan's multi-year Spotify renewal was never just about one show.

Podfade looks different from the advertiser's side, and the broader numbers explain why. Listen Notes counted more than half a million podcasts active in 2026, the most since the pandemic boom of 2020 and 2021, even as new launches fell to their slowest pace since 2017, according to figures reported by Podcast News Daily. Nearly twenty million episodes have been published this year, and nearly three in ten new launches in the third quarter were AI-generated, a trend already bringing AI podcasts into listeners' feeds. Fewer launches, steadier output, more machine-made shows: the industry is maturing into something that looks less like a gold rush and more like a media business with a real casualty list.

So the next time a feed goes silent a month in, the data says that is the ordinary outcome. The podcasts that make it past the early cliff tend to be the ones where a host books a guest, talks for longer than ten minutes, and does not try to publish every single day. Everything else is podfade, and the odds were probably decided by episode three.