Pondering Podcasting’s Partiality Problem
As social media starts to warm up to podcasting, podcasters are excited at the prospect of “going viral.” Will we just be chasing vanity metrics? Or is this a real chance for meaningful show growth?
With millions of podcasts-and tens or hundreds of millions of podcast episodes-now available, the tubes that get your podcasts’ episodes to your listeners are getting a little jammed up. OK, that’s not how the internet works, Senator. But it’s an unfortunately apt metaphor for where podcasting is today.
Seasoned podcasters have learned to expect a delay between time-of-publish and when the latest episode will publicly appear in various directories. At the risk of throwing out another metaphor, when a clothing store gets in a new shipment, it takes a little time before the mannequins in the front window change outfits.
We’re OK with this because we assume that, just like a store getting a new shipment, existing orders will be filled first, and then the showroom will be adjusted. In the case of podcasting, those “existing orders” are people who have chosen to subscribe to/follow our RSS feed in a podcast app of their choosing. Subscribers/followers, we’ve assumed, bypass the public directory update and, we continue to assume, our episodes arrive in our listener’s apps within minutes of publishing.
We’re wrong about that.
Construction Ahead. Expect Delays
Last week, a podcast I’m subscribed to published an episode in the early hours of Thursday morning. No matter how many times I refreshed Apple Podcasts, it wouldn’t deliver for me. On Friday, I got sick of waiting and listened to the episode with Spotify. Saturday evening, the episode finally was available to me. And, again, I am subscribed/following this show in Apple Podcasts.
Coincidentally, I published Thursday’s Podcast Pontification episode precisely at 9:55a Arizona time, as I always do. Yet it didn’t make it to my phone until after 5:00p that evening, an update cycle well in excess of the few minutes I’ve come to expect.
James Cridland of Podnews has an excellent breakdown of the infrastructure change Apple has put in place that exacerbates this problem for podcasting. The headline is this: All podcasts are not treated equally by Apple.
Instead, Apple is determining which shows are worthy of lighting-fast updates, and which podcasts are not worthy. Surprising absolutely no one, podcasts with much larger audiences that publish content on a very frequent basis are deemed by Apple as more worthy of more rapid updates than the rest of us plebes.
This Is The Way
If you are as troubled by this as I was troubled by this when I learned of it, you’ll be more troubled-as I was-to learn this is how just about every podcast directory works. It’s not just Apple.
Personally, I don’t like being troubled by reality. And reality says that treating every RSS feed exactly the same is a waste of resources. Because the majority of the millions of RSS feeds haven’t had a new episode in months, no competent developer is going to write code that checks and parses all podcast RSS feeds every minute of every day. That’s wasting bandwidth and processing power, neither of which are very green.
So, like Apple, podcast directory developers write algorithms to evaluate the worthiness of each show in their catalog based on a variety of factors to determine the optimal update frequency for that show on their platform. If your podcast isn’t deemed as worthy, your episodes will take longer to get onto the devices of your listeners who use that directory.
If that’s knowledge causes you more troubles, I have bad news for you: This is how Google works. Google doesn’t crawl all websites in the world with the same cadence, frequency, or depth. Sites more worthy by Google of frequent crawling are more frequently crawled. For websites deemed less-worthy, any changes that Google discovers during that crawl can take time to populate their directory, causing outdated information to remain on their SERPs (search engine results pages). How long this updated info remains on the SERP is dependent on Google’s worthiness assignment of that site or page.
Coming To Terms With Podcasting’s Preferential Treatment
In hindsight, I think we all knew something like this was happening in podcasting. Many of us didn’t give it too much thought because we wrongly assumed we could bypass the slow lane by getting our listeners to subscribe to/follow our shows.
But we can no longer safely assume that.
Podcasts that publish episodes designed to be consumed during the evening commute, like Techmeme Ride Home or Post Reports by The Washington Post, can no longer safely assume their episodes will reach their listeners’ devices ahead of the commute.
Podcasts that publish episodes designed to be consumed before the workday begins, like the 7am Podcast out of Australia, can no longer assume their episodes will reach their listeners’ devices by, well… 7:00 in the morning.
And it’s not just those shows. Dayparting in podcasting is a growing trend. I read a story (that I can’t find now) about a podcast that publishes eight episodes per day! They can’t be confident that all of their episodes will reach all of their listeners.
Not with a complete dependence on RSS-based distribution via podcast apps, that is.
The Killer App Remains Killer For A Reason
The reason we can’t trust podcast directories to deliver our content in a timely fashion is because it’s not a one-to-one relationship. We think it is, but it’s not. The directory sits in the middle, and they control the terms of the relationship. Not us. Not our listeners.
But email is a one-to-one relationship.
Yeah, I know you don’t want to listen to a podcast in Outlook. Neither do I, so I’m not suggesting you stop publishing an RSS feed or anything resembling that behavior.
Instead, I’m-once again-suggesting you start establishing a one-to-one email-based relationship with as many of your listeners as you can.
When I send out an email to all of the people subscribed to Podcast Pontifications In Your Inbox at 9:55am, there’s no one in the middle controlling the distribution. It may take a few minutes for my email-sending service to churn through my list. And if you have hundreds of thousands of people on your email list, it’ll take some time for the send to complete. But there isn’t a directory to update. And all mail systems are quite transparent in their sending process, allowing senders to adjust their start time as necessary to achieve 100% distribution by a certain time.
So if you never want to leave your audience in suspense, consider email podcast distribution as a backup plan or fail-safe. How in-depth you go in those emails is up to you. I invest quite a bit of effort into the emails that I send. Others do little more than a quick “new episode” notice with a link to the audio file. I’ve a preference, but baby steps are a smart play here. Just start.
If you haven’t been frustrated by slow delivery times, consider yourself lucky. Many podcasters aren’t so lucky. If someone in your podcasting peer group is dealing with this, please send them this article to help them understand why. And also get them thinking about establishing a one-to-one relationship with a subset of their listeners via email.
And if you love the idea, I would love for you to go to BuyMeACoffee.com/evoterra where you can slide a virtual coffee my way.
I shall be back tomorrow with yet another Podcast Pontifications.
Originally published at https://podcastpontifications.com, where it started life as an episode of my four-times-a-week short-form podcast called, oddly enough, Podcast Pontifications. It’s a podcast for working podcasters that’s focused on trends in our growing industry and ideas on ways to make podcasting not just easier, but better. Yes, you should listen. Here’s an easy way: 👇
Evo Terra (hey, that’s me!) has been podcasting since 2004, is the author of Podcasting For Dummies and Expert Podcasting Practices for Dummies, and is the CEO and founder of Simpler.Media, a strategic podcast consultancy working with businesses, brands, and professional service providers all around the world.