The Chapter Mark: How Britain's Biggest Podcasts Are Building a Second Table of Contents

When the timeline learned to talk

Open Apple Podcasts mid-episode of The Rest Is History, swipe up, and a panel appears that wasn't there four years ago. A list. Each row is a moment in the episode — a timestamp, a phrase, occasionally a tiny image. Tap one and you're transported. It's a table of contents, but for a thing that was supposed to refuse them: a piece of audio you were meant to drift through, not navigate.

These are chapter marks. They've been part of the MP3 spec since the early 2000s, supported in fits and starts by Apple Podcasts since iOS 14, and properly enshrined by the Podcasting 2.0 namespace and the PSC tag since around 2021. Pocket Casts and Overcast have backed them for years; Spotify still hasn't, with predictable consequences for production decisions. And in the last eighteen months, British podcasts have started taking them seriously — but in wildly different ways, with the predictable bell curve of brilliance, indifference and small disasters.

What the listener actually sees

Open The Rest Is History episode 487 ("The Norman Conquest, Part 3") in Pocket Casts and you get six chapters. The same episode in Apple Podcasts shows the same six chapters with cover artwork swapped per chapter — Goalhanger ship those too. In Spotify, you get nothing. The show appears as a single seventy-eight-minute brick, scrubbable but blind. The chapter data is in the file; the app just refuses to parse it.

The effect on listening behaviour is measurable, if you trust app-side telemetry. Pocket Casts published a graph in their 2024 listener report showing chapter-tap rates rising sharply for episodes over forty-five minutes. Snipd, the AI-powered listening app, lets users save chapter-bounded clips with one tap, which collapses the whole "I'll come back to this" instinct into a single gesture. For a craft that has spent two decades insisting that audio is linear, the chapter mark is a quiet admission that, actually, sometimes it isn't.

Eight British podcasts and how they chapter

For this piece we sampled three consecutive episodes of eight British shows in April and May 2026, pulled directly from the published feeds and inspected in Pocket Casts and the Podcasting 2.0 index. Counts and patterns reflect what was actually shipped, not what the show says it does.

The first thing that jumps out is the spread. The Diary of a CEO runs about a dozen chapters per ninety-minute episode — almost industrial in its segmentation. Today in Focus and The Infinite Monkey Cage ship none at all. Two of Britain's most-listened-to daily and weekly podcasts, both publicly funded or quasi-funded, both treating their feeds as a single block of audio because their primary platform isn't a chapter-aware app. Monkey Cage is BBC Sounds-first; Today in Focus is built for the Guardian's own player. The chapter spec exists. They've chosen not to use it.

Three things the best ones get right

1. The chapter title is the editorial promise, not the production note. The Rest Is History's chapter titles for the Norman Conquest run read like a Penguin Classics contents page: Senlac Ridge: The Battle Begins; Harold's Wall of Shields; The Feigned Retreat. They're written by someone — almost certainly the producer team — to be browsed. Compare with shows whose chapter titles read like Audition export labels (Section 2 – Mic 1 cleanup). The difference is whether someone gave the chapter a sentence's worth of editorial thought.

2. Ad breaks get their own chapter, and they're labelled. The News Agents and Empire both flag ad chapters explicitly — Ad Break, or on Empire the slightly more elegant A Word From Our Sponsor. This is good manners and good UX: a chapter-aware app like Pocket Casts will let you skip the whole ad pod in one tap. Some American shows (Vox, This American Life) have done this for years; British shows are catching up. The Diary of a CEO labels every break, which is part of why its chapter count is so high — they're using chapters as both navigation and ad-skip metadata.

3. Chapter artwork carries the brand. Goalhanger's flagship shows (Rest Is History, Rest Is Politics, Empire) ship per-chapter cover art. It's the same colour palette as the main feed art, but with a different inset image per chapter. It costs them perhaps an extra forty minutes per episode in design time, almost nothing in file size (chapter art is typically embedded at low resolution), and it makes the show feel hand-bound. The contrast with the wall of unchaptered audio coming out of BBC Sounds is stark.

Three failure modes

1. The chapter at fifty-eight minutes that just says "Sign-off". Some shows seem to chapter as an afterthought — three or four marks in the first twenty minutes and then nothing until the outro. This is worse than no chapters, because it implies a structure that doesn't exist. Listeners scrubbing back to find "that bit about Cromwell" can't, because the chapter that should bracket Cromwell never got cut.

2. Auto-generated chapters from a transcript pass. A small number of shows now run their finished episode through a transcript-and-summarise pipeline that emits chapter titles automatically. The results read like LinkedIn — The Importance of Authenticity in Leadership; Why Failure is Necessary. Listeners can tell. The clue is when every chapter title contains a noun phrase no human host would say aloud.

3. Chapters that fight the show. Off Menu's menu-item chaptering (Starter, Main Course, Dessert, Drink, Side Dish) is conceptually clever and practically brittle: when the format breaks — which it sometimes does for guest reasons — the chapter titles lie. A chapter labelled Dessert that contains a tangent about cricket isn't a chapter, it's a misfile.

A short anatomy

For producers wondering what to actually ship, here's what the best British examples have settled on:

1. Five to seven chapters per hour of episode. Below five and the navigation isn't useful; above eight and you're chaptering paragraphs.
2. One sentence or less per chapter title. Senlac Ridge: The Battle Begins — six words. Anything longer truncates in Apple Podcasts' chapter list.
3. First chapter starts at 00:00. Not at the end of the cold open. This sounds obvious, but a surprising number of shows start their first chapter mark thirty seconds in, leaving the cold open uncategorised and ambiguous.
4. Ad breaks always chaptered, never silently dropped in. Listeners will find out anyway; doing it openly buys goodwill.
5. Chapter art is optional but it's the cheapest upgrade in podcasting if you have a designer on the team.

Where the chapter mark is going next

Two things are about to shift this. The first is Spotify; the company has been telling production partners since late 2025 that chapter support is "in development" — which historically means it ships when Spotify needs another announcement, but it does seem to be coming. Once Spotify joins the party, every flagship show will chapter, because the cost of not doing it becomes a missing feature in the dominant player.

The second is the listener-side AI tools we wrote about in our piece on the Listener's AI Toolbelt. Apps like Snipd are already doing automatic chapter detection on shows that don't ship their own. The result is that the chapter mark is becoming an editorial choice with a fallback: either you do it yourself and own the framing, or someone else's algorithm does it for you and chooses the framing for your listeners. Goalhanger has noticed. The BBC, on current evidence, has not.

For a publication that obsesses over the craft details that listeners notice without being able to name, the chapter mark is one of the cleanest examples of the genre. It doesn't change what the show sounds like. It changes what the show is willing to admit it is: a structured thing, navigable, browseable, made of parts. That's a small philosophical shift, dressed up as a metadata field. The best British shows are getting there. The rest are about to be dragged.