Podcast chapters — the table of contents your episodes are missing

Why chapters matter in 2026

The median top-100 episode now runs past 60 minutes. Nobody has a free hour, and a listener who can't find the segment they came for either scrubs or leaves. Chapters fix both: tapping a titled segment jumps straight to it, and — counterintuitively — giving permission to skip increases total listening time, because people press play when they know they can jump.

The SEO angle is the overlooked one. Apple Podcasts indexes chapter titles and uses them to surface episodes in in-app search — descriptive titles (guest names, company names, specific questions) become searchable terms. Spotify similarly surfaces chapter-level content in search and recommendations, and episodes with chapters generate richer engagement signals that feed its recommendation engine. On the web, each chapter title is another indexed text element on your episode page, and chapters make clip selection for social trivial — every chapter is a candidate reel.

Despite all of this, a 2025 analysis of the top 500 Apple Podcasts shows fewer than 30% ship chapters. The reason is honest: scrubbing a finished 60-minute episode to place each marker manually is an hour of unpaid work per episode. That's a transcription problem, not a creativity problem — and it's solved below.

The three formats — and who honors what

Verified against the Podcasting 2.0 namespace spec, Podlove spec, Apple's guidelines, and 2026 app behavior.

The 2026 minimum: description timestamps for universal reach, plus the podcast:chapters JSON file for Apple and Podcasting 2.0 apps if your host supports it. ID3 is optional garnish — skip it if your host strips tags.

The podcast:chapters tag, copy-paste

The tag is a pointer, not a container — it links to an external JSON file. Two parts: declare the podcast namespace once on the <rss> element (you already did if you ship podcast:transcript), then add one tag per episode.

<rss version="2.0"
     xmlns:podcast="https://podcastindex.org/namespace/1.0">
  <channel>
    <item>
      <title>Episode 42: The Future of Live Audio</title>
      <enclosure url="https://example.com/episodes/42.mp3"
                 length="38212000" type="audio/mpeg"/>
      <podcast:transcript url="https://example.com/transcripts/42.vtt" type="text/vtt"/>
      <podcast:chapters url="https://example.com/transcripts/42-chapters.json"
                        type="application/json+chapters"/>
    </item>
  </channel>
</rss>

The chapters file itself — startTime is in seconds:

[
  { "startTime": 0,   "title": "Intro" },
  { "startTime": 128, "title": "Interview begins", "url": "https://guest.example" },
  { "startTime": 905, "title": "Why most podcasters quit in year one" },
  { "startTime": 2210, "title": "Closing recommendations" }
]

Because chapters are a separate file, you can fix or extend them after publishing without re-uploading audio — impossible with ID3. JSON chapters also work in web browsers (which can't read ID3), and per-chapter images load on playback instead of bloating the audio file. The type attribute is required; the spec prefers application/json+chapters.

Apple's acceptance rules

Apple displays your chapters only if they meet its guidelines — otherwise listeners on up-to-date apps get Apple's generic auto-chapters instead of yours:

Density guidance beyond Apple's floor: a break every 4–8 minutes — 3–6 chapters for a 15–25 minute episode, 6–10 for 30–50 minutes, 10–18 for 60–90 minutes, 15–25 for 90+. Too few and listeners still scrub; too many and the list becomes noise. And name chapters for humans and search: "Introduction" adds nothing; "Why most podcasters quit in year one" tells the listener and the in-app search index exactly what's there.

Generate chapters from a transcript — free, local

The manual cost is what kills chapters: an hour of scrubbing per episode. But a timestamped transcript already contains every chapter you need — the topic shifts are the chapter boundaries, and the SRT timestamps are the chapter times. Cloud generators charge for exactly this (Castmagic, Podsqueeze, and WhisperTranscribe all gate chapters behind paid tiers); the local workflow produces the same list free:

The timestamps you already produced for captions are the exact data chapters consume — which is why the transcription step, done locally, is the whole cost. Our guide to local podcast transcription covers the transcription pass; the podcast:transcript tag covers linking the files from your feed.

Related reading

Podcast chapters — FAQ