Podcasters have a lot of freedom: no gatekeepers, fewer rules, and no broadcast schedule dictating how long or short each episode needs to be. As a result, podcasts range from as short as a minute to as long as 3+ hours. But what do listeners really want? And can tweaking your episode length get you more downloads?
Listening Data Disclaimer
Aggregate podcast listening data is still hard to come by, since publishing and listening are so decentralized. In many cases we don’t even know if a user actually listened after downloading. In other cases, app makers and hosting providers collect detailed listener data, but won’t share that information publicly, or even with the podcasters themselves. All that to say that the data we’re about to discuss should be taken with a grain of salt. With that said, let’s see if we can infer anything from the available information.
Episode Length vs. Time Spent Listening
One data point comes from podcast hosting service Omny Studio. In 2016 they released this graph as part of their “Beyond the Download” webinar.
It shows how podcast length relates to “time spent listening” or TSL. According to this the average listener gets through 62% of a podcast under 5 minutes long versus only 22% of episodes over an hour. The data is flawed - it was collected from an unknown sample size of Omny Studio hosted shows - but it does provide some food for thought.
Takeaway: Shorter might be better - and content that appears earlier in an episode is more likely to be heard.
A lot of listeners don’t make it to the end of the show - so frontload your most important content. Ads and other content after the first quarter of the show may not be heard, especially in longer shows.
But Long Shows Are Popular, Aren’t They?
At the 2017 NAB Show, LibSyn VP Rob Walch showed that 84% of the biggest podcasts (those with more than 100,000 downloads) were more than 51 minutes long - while less than 10% of those popular shows were under 30 minutes. Walch says “Podcast consumers want to hit play and don’t want to have to fumble around with the device again”. This makes intuitive sense, and LibSyn is one of the biggest hosting companies, with access to a huge volume of data - but there are major flaws with this research too.
First, we’re looking at downloads, not listens, so we can’t tell what percentage of those 51+ minute shows are actually being heard. Plus, there’s the question of causality: are these podcasts popular because they’re long? Long because they’re popular? Are the two related at all? We can’t tell based on LibSyn’s data.
Takeaway: A lot of highly-downloaded podcasts happen to be long, but it’s unclear if that’s because consumers want longer shows, or if other factors are at play.
More Variety Means Shorter Attention Spans
Circling back to time spent listening, Bridge Ratings’ 2017 Podcasting Best Practices Study found a drop in TSL in recent years.
By studying a sample of 2000 people, they found that individual listening sessions went from an average of 45 minutes in 2015 to 22 minutes in 2017. At the same time the average number of podcasts listened went from 4.5 to 7. The authors did note that 56% of those surveyed listened to podcasts in multiple sessions - meaning they continued listening to episodes beyond a single session.
Takeaway: Podcast listeners have more choice than ever - which means they’re less likely to stick with you for long stretches of time. If your episode is longer than about 22 minutes, it better be compelling enough draw listeners back for a second listening session.
No Matter the Length, Start Strong
2016 data from NPR One shows typical podcasts lose 20-35% of listeners within the first 5 minutes. In fact, the drop-off rate is higher in the first 5 minutes than at any other time during the episode. Just look at the steep drop in the first two minutes in this graph of listener retention. It’s worth noting too the steady drop off of listeners throughout the entire 16 minutes shown. That slope is another argument for shorter shows if you ask me.
NPR goes on to say their data also suggests that 18-30 minutes may be a “sweet spot” for keeping an audience engaged. NPR isn’t saying much about its methodology here, but we can assume they’re working from a large, if NPR-centric, data set.
Takeaway: Make the first 5 minutes of your show count - or risk losing listeners. Get to the point quickly and structure your content to keep your audience engaged. Episodes 18-30 minutes long may be ideal, at least for NPR One listeners.
So... How Long Should My Podcast Be?
As you may have guessed by now, there isn’t an ideal length. Yes, there’s some evidence that shorter is better, but many long shows are thriving. Just look at the Apple Podcasts top 50 chart:
Yes, I’m adding to the poor quality data out there by creating this highly unscientific image - but I think it illustrates a couple of things. Just look at how many long podcasts there are in the top 50! By my count 70% have at least one episode over 40 minutes long in their latest five, 26% stay in the 20-40 minute range, and just 4% were shorter than 20 minutes.
Like LibSyn’s data mentioned above, it’s hard to discern causality. But looking at this set I notice a few things that might explain why so many popular shows are also long:
- Popular shows from NPR and the like are long because they’re built for the radio broadcast schedule first.
- Celebrity-driven shows like Joe Rogan’s might have a higher tolerance for longness thanks to the rabid fanbases of the host and celebrity guests.
- Best in class production companies like Gimlet have the skills and budgets to make long shows compelling (and the promotional budgets to move them up the charts).
So unless you fall into one of the categories above, it may not be best to follow in the footsteps of popular shows that also happen to be long. For most podcasters it’s all about finding a balance between length, ease of production, and compelling content - like the classic “good, cheap, or fast - pick two”:
Conclusion? Make Your Show However Long It Deserves To Be
Creating compelling audio content isn’t easy, and making something that’s compelling for over an hour or even 30 minutes is even tougher. Audiences have an embarrassment of choice in the podcast landscape today, and it seems they’re increasingly likely to “channel surf” - abandoning shows that don’t hold their interest for ones that do. Legacy media companies, big budget startups, and celebrities have the skills and cachet to hold audience attention for longer periods, and the marketing capacity to top the charts. There’s no evidence that longer shows are preferred by listeners in general though, in fact there appears to be a growing appetite for shorter episodes.
Takeaway: When in doubt, go for short and good over long and boring (duh). If you’re an indie host or producer, don’t imitate the big guys when it comes to length - they may be the exception rather than the rule.
One last thing. Short shows can be more work than long ones. Editing a 45 minute interview down to 10 minutes of gold isn’t easy, and writing concisely can be time consuming. Mark Twain said: “I didn't have time to write a short letter, so I wrote a long one instead.” And isn’t it better to send that long letter than nothing at all? So I say make the show you want to make, and one that you’re capable of making on a regular basis. Your goals may be best served by Don’t get bogged down worrying about length - because in the end wouldn’t you rather lose some listeners halfway through than have no listeners at all?