How long should a podcast episode be? What the hosting data shows

Thirty-one percent of the episodes hosted on Buzzsprout run between 20 and 40 minutes, according to the platform statistics Buzzsprout publishes for September 2026. So how long should a podcast episode be? Twenty to 40 minutes is the honest default, because that is where the largest block of working shows already sits, and a target inside that band will not cost you anything. The exceptions are narrow and they are set by format rather than taste: interview shows run long, daily news runs very short, and no platform publishes evidence that one length wins listeners over another.

Before the detail, one practical note. Any podcast can be listed free on PodAnswer, and a listed show gets its own page that can surface when somebody searches the show's name, so if yours is not on the site yet you can add it at podanswer.com/signup.

The spread matters more than the middle

Buzzsprout's public platform stats page breaks every episode it hosts into five duration buckets. For September 2026 the split was: less than 10 minutes, 20 percent; 10 to 20 minutes, 16 percent; 20 to 40 minutes, 31 percent; 40 to 60 minutes, 20 percent; over 60 minutes, 13 percent. The company states on the same page that the figures cover 111,813 active podcasts and 148,862 new episodes that month.

Read that distribution carefully, because it does not say what most length advice claims it says. Thirty-six percent of episodes are under 20 minutes. Thirty-three percent are over 40. The 20-to-40 band is the plurality, not the majority, and the second-largest single bucket is the shortest one on the chart. A show running eight minutes an episode and a show running 70 are both sitting inside a large, established group of peers.

Buzzsprout is one host, not the whole industry, and it skews toward independent shows rather than network productions. Treat the numbers as a large sample of the market you are actually competing in rather than a census.

What the tag in your feed will accept

Episode length is published to the apps through a single RSS element, itunes:duration. The podcast de facto standard documentation at podcast-standard.org, which bases its figures on a sample of 258,607 podcasts, describes it this way: duration "Can be in seconds or in the format hh:mm:ss. Duration in seconds is recommended." The same reference reports the tag appearing in 97.2 percent of feeds, which makes it one of the most reliably populated fields in podcasting.

Apple's own technical rules say nothing about how long an episode may be. The Apple Podcasts for Creators page on podcast RSS feed requirements covers the XML declaration, the enclosure tag with its "three required components, URL, length, and type", the globally unique identifier that "never changes", ASCII filenames, RFC 2822 dates and character encoding. There is no duration ceiling anywhere on it. The one requirement with a bearing on long files is this instruction: "Display your show artwork and allow streaming playback of your episodes by enabling your hosting server for HTTP HEAD requests and byte-range requests." A 90-minute episode on a host that does not support byte-range requests is an episode listeners cannot scrub through, which is a hosting problem wearing a length problem's clothes.

The number you are really choosing

Length is not an isolated decision. It trades directly against how often you publish, and Buzzsprout publishes that distribution too. In September 2026, 7 percent of shows released an episode every zero to two days, 35 percent every three to seven days, 36 percent every eight to 14 days, 20 percent every 15 to 29 days, and 2 percent less often than monthly.

Put the two charts side by side and the real constraint appears. Weekly publishing plus 45-minute episodes is roughly 39 hours of finished audio a year, and for a one-person show the editing behind that is the job, not a side effect of it. Most hosts who burn out do not burn out on recording. They burn out on the back half of a long edit, every week, for a year. Cutting a target from 45 minutes to 25 buys back the time that keeps the schedule intact, and schedule reliability is the thing a listener notices.

If you have already published a back catalogue at an unsustainable length, the archive is still worth something. We covered what to do with old episodes nobody plays in the back catalog piece.

Three formats, three defensible targets

The useful version of this question is per format, not per industry.

Interview shows. Plan for 40 to 60 minutes and edit toward the lower end. A guest conversation has a natural shape of roughly an hour, and a hard cut at 35 minutes usually means losing the part where the guest stops performing. Twenty percent of Buzzsprout episodes land in the 40-to-60 bucket, so this is well-populated ground.

Solo and commentary shows. Fifteen to 25 minutes. One person talking has a shorter half-life than two people talking, and the 36 percent of episodes that run under 20 minutes are disproportionately this format. A solo episode that runs 50 minutes is almost always a 20-minute episode with three tangents in it.

Narrative and documentary. Length follows the story, typically 25 to 45 minutes per episode, with the real discipline applied at the season level rather than the episode level. These shows also carry the heaviest production cost per finished minute, which is why serial seasons publish in runs and then stop.

Daily news and briefings. Under 10 minutes, and ideally under six. This is the format driving a good share of that 20 percent sub-10-minute bucket, and the format where a listener's tolerance for padding is lowest.

Testing it on your own feed rather than on somebody else's average

Twenty episodes in, your own analytics beat any benchmark. Apple Podcasts Connect and Spotify for Creators both report per-episode retention inside their own dashboards, and that curve is the only length evidence specific to your show. Two things are worth looking for. First, where the drop-off cliff sits: if a consistent share of listeners leaves at the 12-minute mark across many episodes, something structural happens at minute 12. Second, whether your longest episodes retain proportionally or absolutely fewer listeners, because those are different problems with different fixes.

What that data will not tell you is whether length changed your download numbers, since the two platforms count differently and neither reports across the other. For context on what download figures mean in the first place, our downloads benchmark piece goes through the percentiles.

Length and the sponsor conversation

Longer episodes create more ad inventory, and that is the one mechanical argument for running long. A 45-minute episode holds a pre-roll, a mid-roll and a post-roll comfortably. A nine-minute episode holds a pre-roll and not much else.

What length does not do is change the audience number a sponsor asks about. Buzzsprout's September 2026 figures for episode downloads in the first seven days after release put the median podcast at 27 downloads, the top 25 percent at 96, the top 10 percent at 412, the top 5 percent at 1,007 and the top 1 percent at 4,586. Buzzsprout states that all of its statistics "comply with the IAB Podcast Measurement Technical Guidelines 2.2". Those thresholds are set by audience size, and adding 15 minutes to an episode does not move a show from the median to the top quartile. Our guide to landing sponsors without a huge download number deals with the pitch side of that.

Search is the other reason to care about the shape of an episode rather than its raw length. A 60-minute conversation that answers six distinct questions gives you six things to publish and index, which is the argument made in our podcast SEO guide and in the piece on transcripts and SEO. A 60-minute conversation that circles one topic gives you one.

What nobody publishes

No platform publishes completion rates by episode length. Apple does not. Spotify does not. The hosting companies that could compute a decay curve across their whole catalogue do not publish one either. Every claim you will read that a specific number of minutes is optimal for retention is extrapolated from somebody's own show or from a survey, not from platform data, and it should be read that way.

The chapter markers that would let listeners skip within a long episode are also far less common than the advice suggests. podcast-standard.org reports the psc:chapters tag in 2.7 percent of the feeds in its sample, and the podcast:transcript tag in 6.3 percent. Long episodes mostly ship without navigation.

Frequently asked

Is there a maximum episode length? Apple's published RSS requirements set none. Practical limits come from your host's file size or storage caps, which vary by plan, so check your own host's documentation rather than Apple's.

Does a longer episode rank better in the podcast apps? No published ranking input includes duration. Apple's own search documentation lists metadata, popularity and listener behavior, and episode titles are the episode-level field it indexes.

Should every episode be the same length? No, but the variance should be deliberate. A show that swings between nine and 75 minutes with no signal in the title teaches listeners they cannot predict what they are downloading. Put the length cue in the title or description, which is covered in our piece on show notes and the feed description field.

How long should a first episode be? Shorter than your target. A 15-minute first episode that exists beats a 45-minute first episode that is still being edited, and the first-year growth plan explains why the first ten episodes matter more as a habit than as a catalogue.

Does YouTube change the answer? It changes the packaging more than the length. Our walkthrough on putting a podcast on YouTube covers the RSS route and what happens to the audio there.

Getting the episode found after you publish it

Pick a target length, hold it for ten episodes, then read your own retention curve and adjust once. That beats any benchmark, including the one at the top of this piece.

If you want the questions inside your episodes working as search pages, that is what PodAnswer does: we pull the answers out of your transcripts and publish them as articles that link back to the episode. The Growth plan is $149 a month for four articles and the Network plan is $349 a month for twelve. How it works for podcasters and the plan details are both on the site, and a free listing is still the place to start.

Sources: Buzzsprout Platform Stats for September 2026, published at buzzsprout.com, for the episode duration and publishing frequency distributions, the first-seven-day download percentiles, the active podcast and new episode counts, and the IAB Tech Lab compliance statement. Apple Podcasts for Creators, podcast RSS feed requirements, for the enclosure, GUID, date and byte-range requirements and for the absence of any duration limit. The podcast de facto standard documentation at podcast-standard.org for the itunes:duration definition and the tag usage figures across its 258,607-podcast sample.