The Correction Episode: How British Podcasts Handle Mistakes (and Why the Medium Has No Ombudsman)

The Structural Gap

In October 2025, a presenter on one of Britain's most-listened-to daily news podcasts stated, with full conviction, that a minister had resigned the previous evening. The minister had not. The episode was downloaded roughly 180,000 times before anyone raised a flag. The team spotted the error themselves, re-exported the segment, and pushed a corrected file to their hosting platform inside twenty minutes. No note was appended. No social-media correction was posted. The episode simply changed.

If the same error had aired on BBC Radio 4's Today programme, it would have triggered a formal correction on the next bulletin, a note in the programme's online transcript, and — depending on severity — a published finding from the BBC's Executive Complaints Unit. If it had appeared in The Guardian, it would have generated a prominent correction clause at the foot of the article, timestamped and immovable.

Podcasting sits between these worlds and inherits neither of their accountability machinery. It is not regulated by Ofcom in the way broadcast radio is (outside the BBC's own podcasts, which fall under the BBC Charter). It has no equivalent of IPSO or Impress, the UK's press regulators. And because podcast episodes are distributed as static files via RSS — downloaded once, cached locally, and never updated on the listener's device — a correction published to the feed has no mechanism to reach the ears that already heard the wrong version.

This is not a niche concern. An estimated 42 million podcast episodes are downloaded in the UK each week, according to Rajar's Spring 2025 MIDAS survey. A growing share of those listeners treat podcasts as their primary news source. Yet the medium's corrections infrastructure remains essentially voluntary, invisible, and — in many cases — non-existent.

The Spectrum of Response

Broadly, British podcasts fall into one of five categories when they need to correct a mistake. The spectrum runs from institutional-grade transparency to what can only be described as hoping nobody noticed.

1. The Institutional Model: BBC Podcasts

The BBC's podcast output — from Newscast to The Today Podcast to In Our Time — sits inside the corporation's broader complaints and corrections framework. Listeners can submit complaints through the BBC website. If upheld, corrections appear on the BBC's Corrections and Clarifications page and, where appropriate, are read on air. The BBC publishes roughly 250–300 upheld complaints annually across all output; a small but real fraction of those involve its podcast-only content.

What distinguishes the BBC model is not that it never makes mistakes — it does, frequently — but that there is a documented, searchable record when it does. The correction is not ephemeral. It exists as a permanent public artefact, linked to the original output, discoverable by anyone who cares to look.

2. The Professionalised Model: Large Independents

Goalhanger, the production company behind The Rest Is Politics, The Rest Is History, and The Rest Is Football, operates what you might call a soft-corrections regime. Presenters will occasionally address errors at the top of the following episode — "a couple of you pointed out that I said 1847 when I meant 1848, and you're absolutely right" — a technique borrowed from live radio but adapted for the asynchronous podcast audience. These corrections are warm, conversational, and effective at maintaining trust with regular listeners. They are also entirely unsearchable. Unless a listener happens to hear the correction in the subsequent episode, the error stands as the public record.

Goalhanger does not maintain a public corrections page. It does not append notes to show notes when factual errors are identified. The correction exists as audio, and only as audio, inside an episode that may have a wholly different title and topic.

3. The Silent Edit

This is the most common approach among mid-tier independent podcasts, and it is worth dwelling on because it raises the deepest questions about the medium's relationship with truth. The workflow is straightforward: an error is identified (by a listener, a guest, or internal review), the offending segment is cut or re-recorded, the audio file is re-exported, and the new version replaces the old one on the hosting platform. The RSS feed's <enclosure> URL stays the same. The episode's publication date stays the same. No public note acknowledges the change.

From the producer's perspective, this is efficient. The corrected version is what future listeners will hear. From the audience's perspective, however, the 180,000 people who downloaded the original have no way of knowing they heard a version that contained a factual error. Some of them will repeat it. Some will act on it. And if the error was about a person — a misattributed quote, a misstated allegation — the subject of the error may never know it was corrected, because the correction made no sound.

4. The Dedicated Correction Segment

A small number of British podcasts have built correction into their format. The Bugle, the long-running satirical news podcast hosted by Andy Zaltzman, has made corrections a running bit — Zaltzman reads listener corrections in a deliberately mournful tone, treating each as a personal failure to be savoured. No Such Thing As A Fish, from the QI team, treats corrections as bonus content, often funnier than the original fact. Page 94, the Private Eye podcast, simply publishes a regular "Errata" episode, separate from the main feed, collecting corrections from across their print and audio output.

These examples share a common thread: they turn correction into content, which aligns the editorial incentive (get it right) with the commercial incentive (produce engaging material). It is a neat trick, and it works precisely because the podcast format rewards personality and self-deprecation in a way that institutional corrections pages do not.

5. The Nothing

The least defensible but most common response: do nothing. No correction, no edit, no acknowledgment. The episode stays in the feed as though the error never happened. This is not limited to amateur productions. In early 2025, a well-known British interview podcast aired a guest's claim that a public figure had been "charged" with an offence. The public figure had not been charged; they had been interviewed under caution, which carries no legal finding. The distinction matters enormously under UK defamation law. The episode remains in the feed, unchanged, at time of writing.

Comparison: How Six British Podcast Outlets Handle Corrections

Why the Medium Makes It Hard

Podcasting's corrections problem is not, for the most part, a problem of bad faith. It is a structural problem that arises from three features of the medium's architecture.

First, the distribution model. Podcast episodes are delivered as static MP3 or AAC files via RSS. Once downloaded, they live on the listener's device. There is no push mechanism to update them. This means a correction published to the feed — even one prominently placed in show notes or a new intro — reaches only future listeners. The existing audience keeps the wrong version. This is fundamentally different from the web, where refreshing a page serves the corrected copy, or broadcast, where the correction airs to the same audience that heard the original.

Second, the absence of a central record. A newspaper correction is attached to the article it corrects. A broadcast correction is logged against a specific transmission time and date. A podcast correction, if it exists at all, typically sits inside audio that is indexed by episode title and topic — neither of which will mention the correction. There is no podcast equivalent of a media corrections database. Even diligent listeners cannot easily check if a specific claim has been corrected unless they happen to hear the follow-up.

Third, the economic structure. Most independent podcasts operate with small teams and tight production schedules. There is no legal department, no standards editor, no compliance officer. The person who would write the correction is often the same person who made the error — and they have the next episode to deliver. The institutional infrastructure that supports accountability at newspapers and broadcasters simply does not exist at the scale of a four-person production team.

What Might Help

None of the following exists at meaningful scale yet, but each represents a thread that could be pulled.

Standardised correction metadata. The RSS 2.0 specification is essentially frozen, but podcast namespace extensions — such as those developed by the Podcast Index project — could theoretically support a <correction> tag linking an episode to a correction notice. Apple and Spotify would need to render it in their apps for it to matter, and neither has shown appetite for doing so. Still, the technical path exists.

Transcription as an accountability surface. As more podcasts publish transcripts (a topic we explored at The Transcript Shelf), those transcripts become a natural place to append corrections. A reader who searches a transcript for a name or figure can find both the claim and the correction in the same document. Several British podcasts — including Tortoise and The Slow Newscast — already do this.

The listener as fact-checking layer. This is already happening, for better and worse. Listeners on Reddit, Discord, and X regularly compile corrections to podcast claims. The Rest Is History subreddit maintains an informal errata thread. The problem is that this crowd-sourced correction layer is fragmented across platforms and invisible to the majority of listeners who are not in those communities. It is accountability by accident, not design.

Ofcom expansion. The most direct regulatory fix — extending Ofcom's broadcast standards remit to cover on-demand audio — is also the most politically contentious. Ofcom already regulates BBC iPlayer and other video-on-demand services under the Broadcasting Code. Extending that framework to podcasts would bring formal standards for accuracy and impartiality, along with a complaints mechanism. It would also raise immediate concerns about chilling effects on independent speech. The government's 2025 consultation on the future of audio regulation explicitly declined to recommend this step, but the question is not going away.

The Listener's Part

For now, the most meaningful pressure point is the audience. Listeners who email a correction and hear it acknowledged on air are participating in a version of editorial accountability that, however imperfect, is more responsive than most institutional processes. When the team behind The Rest Is Politics addresses a misstatement in the following episode, the correction cycle is measured in hours, not the weeks a press regulator might take to adjudicate a complaint.

The catch is visibility. The correction disappears into the archive, unsearchable and untraceable. The medium gains speed and warmth but loses permanence. Whether that trade-off is acceptable depends on what you think a correction is for — to set the record straight for the individuals affected, or to create a durable public account of what is true. Right now, British podcasting is doing a decent job of the first and almost none of the second.

That the medium has grown this large without a formal corrections infrastructure is, in its own way, remarkable. It suggests that audiences either do not expect accountability from podcasts in the way they do from broadcasters and newspapers, or that they have not yet recognised that the gap exists. Neither interpretation is especially comforting.