Podcasters Have Had Apple's New Podcast Analytics for Two Months Now. What Have They Learned?
Welcome to The Yard, a blog by Backyard Media that explains the podcast industry and podcast advertising.
This is Part 2 of a two-part post about Apple's new Podcast Analytics tool. Read Part 1 here first to learn about the technology.
It was a highly-anticipated bit of news. Apple had been promising podcast publishers and advertisers for a while that it would make Apple Podcasts, still the main pipe through which a majority of listeners get their podcasts, a lot smarter. And at the tail-end of 2017, Apple finally unveiled a beta version of its Podcast Analytics tool. Any podcasters that published their show on iTunes could view statistics about how listeners consumed their episodes once they downloaded them. What was once a keyhole-sized view into the listening habits of podcast audiences suddenly became a wide open window. So, what have we learned from it these last two months?
Before we get into the findings, a few caveats about the data we have so far. First, Apple's Podcast Analytics tool is still very much in beta, and that likely will remain so for another few months. Second, the tool only measures podcast listening on devices running iOS 11. Obviously, this excludes a sizable minority of listeners using Android phones, as well as those listening on feeds that don't rely on iTunes, like Stitcher or SoundCloud. Third, while the number of iPhones running iOS 11 is already a majority (outlets reported in November 2017 that more than 52% of users had upgraded to the new software), there are still some listeners running iOS 10 that Apple's tool can't see. As time goes on, more devices will be running iOS 11 and therefore show up in the Analytics data. Finally, podcasters who use dynamic ad insertion will find their data is a bit muddied. This is because dynamic ads can be different lengths, causing the overall episode length to vary depending on which ad is served to the listener. This variation will affect metrics like total minutes spent listening and percentage of the episode completed.
With that in mind, let's dig into what we've heard so far from podcasters.
Listeners really do listen to the entire episode.
Recall from our blog "What We Know About Podcast Listeners" that self-report studies found that 40% of listeners listened to the entire episode once they start it, and another 45% listened to "most of the podcast" (e.g. more than half of the episode). Some in the podcasting industry were worried that once the "black box" of Apple's podcast data opened up, the actual completion rates wouldn't be as high.
Well, it turns out that podcast listeners' completion rates actually are that high. Many podcast networks like Panoply and Headgum are reporting that their podcast listeners consume on average 80% to 90% of podcast episodes. This means that drop-off is pretty uncommon. Once listeners start an episode, they're engaged enough to finish it. This data matches what we know about podcast listeners - that they exhibit unusually high engagement for a digital audience.
Much of this is due to the "intimacy effect" of podcasts. This comes from listeners hearing hosts speak into their ears for hours on ending; they feel they've developed a relationship with the host and care about the content of the show. We've also discussed in previous blogs that listeners come to podcasts specifically because they like the on-demand nature of the technology and believe they can't get the same content anywhere else. All of this taken together, it makes sense why listeners complete podcast episodes at such high rates - they care about the podcasts they're listening to and actively choose to tune in to them. As Headgum's CTO Andrew Pile puts it, "I think people are overall very relieved to see that people are actually listening the way that we hoped...There are really audiences out there who listen to every word that comes out of [a host’s] mouth."
So, what does this mean for ads? Well...
Ad-skipping isn’t very prevalent.
The 80-90% number above indicates that most listeners actually don't fast forward through the ads. For those listeners who do skip ads, podcasters are finding that those listeners remain engaged with the content of the episode.
They know this by looking at Podcast Analytics' episode-level data, which shows at what time listeners stopped an episode. At the times where mid-roll ads run (those that play in between episode segments), they aren't seeing significant drop-offs. So, even when listeners hear something different, they still stick with the show until the regular content comes back. This makes sense, given that we know that listeners feel that ads are part of the content of the show.
This was another one of the big concerns about opening the "black box" of listener-side data, that a substantial portion of listeners are fast-forwarding through ads. The fact that listeners really are engaged during ads means that download numbers do reflect the number of ads actually heard, and that the value of podcast ads really is as high as we thought. We expect that, because podcasters can now definitively prove the value of their audiences, new sponsors will come into the podcast advertising market.
There is no “perfect length” for a podcast episode, and podcasters should continue to experiment.
Another concern among many podcasters anticipating Apple's release was "over-optimization". That is, that the existence of a new tool pointing to a "correct" podcast length would lead podcasts to optimize uniformly toward one type of podcast format, much like how YouTube’s algorithm changes have led to many creators making longer videos. Podcasting has thrived in the past decade by experimenting with different lengths, sound designs, and narrative formats.
Podcast creators are reporting that they don't see listener completion and drop-off rates changing substantially, even when they change the length or format of their episodes. This is where the Podcast Analytics tool will likely be the most useful for podcasters moving forward. They can make creative decisions about their shows and test them. If there's a negative impact on listenership, podcasters can course-correct. This can help producers fine-tune what works best for their audience, and allow for more diversity in formats. But if the data from the past two months tells us anything, we expect that listeners will remain open to innovation in the type of content and formats they hear from their favorite shows.
Now, podcast sponsors can know for sure how many people heard their ad.
The most exciting part about this new data is that it confirms what podcasters felt they already knew - that their audiences are loyal and sophisticated consumers. Furthermore, they're engaged throughout the episode, and they don't mind advertising, so long as they feel the host is making genuine product recommendations to them. Anecdotally and from self-reporting, podcasters knew that listeners consumed the entirety of podcast episodes once they started them. Apple's new data solidifies that claim.
For established podcast sponsors, the data is a validation that their marketing efforts are reaching listeners. For potential sponsors who have been so far unsure about podcast advertising, Apple's new tool gives them assurances that this medium is valuable and an excellent opportunity to reach new audiences. Even more than that, it lets podcasters inform sponsors how popular their shows really are, down to the performance of specific episodes. This allows for more detailed discussions about CPM rates for different ad placements and for episodes with particular types of content.
As we get even more data from Apple in the coming months, we'll be sure to check back in and share our insights about this important development in the podcasting space.
Reach valuable, engaged audiences with podcast advertising. Contact us to learn how easy it is to get started advertising with Backyard Media's podcast partners.
Backyard Media is a marketplace for podcast advertising. We connect content creators of all shapes and sizes with awesome sponsors, providing them with the resources they need to do what they do best. Everyone wins.
Take a look at some of our related content below:
- What Apple's New Podcast Analytics Tool Tells Us
- What does it mean to be a podcast content leader?
- What does smart speaker technology mean for the future of podcasting?
- When Will Dynamic Ad Insertion Take Over Podcasting?
- Why Listeners Respond to Podcast Ads
- What We Know about Podcast Listeners
- What Does a Podcast Ad Campaign Look Like?
Sources for this post:
Podcast Listeners Really Are the Holy Grail Advertisers Hoped They'd Be by Miranda Katz (Wired).
Turns out people really like podcasts after all (and now we have numbers to prove it) by Nick Quah (Hot Pod/NiemanLab)
Apple launches its podcast analytics service into beta by Sarah Perez (TechCrunch)