What I didn't know about podcast hosting platforms

This article was originally shared on Substack.

So that we understand each other, there was an awful lot. There still is.

When I first started STEAM Powered, I was running on a shoestring, and by ‘shoestring’, I mean (effectively) free.

‘Free’ was also the only criterion I had in choosing a hosting platform and besides, I had no idea what else I should be looking for in a platform anyway. Helpfully, I knew another tech-savvy person who was vlogging/podcasting and they used Anchor.fm. Decision made.

Really detailed research, I know.

Getting started with Anchor

There are mixed feelings about Anchor as a platform based on what I’ve heard from other podcasters,1 but from personal experience, it’s a great starting service. The user interface is intuitive, the stats are basic but give you essential information, and it just works. I was not left wanting in my podcast release workflow as a new creator.

Anchor is great if you:

  • are starting out
  • aren’t prepared to pay for a service
  • don’t want to be limited by the number of episodes you can release before you need to upgrade to a paid plan
  • are only minimally concerned about analytics and monetisation
  • want to release a video podcast (more on this later)

I highly recommend it as a basic or starter service.

You may outgrow Anchor if you:

  • want more detailed analytics or insights (built-in or through a third-party service)
  • are not based in the US and in-app monetisation opportunities are important to you (at the time I left Anchor this was US only, but things may have changed since)

Third-party tracking prefixes

Around a year and a half into podcasting, I joined OSSA Collective, a women+ podcasting community that connects creators with earning opportunities.2

One of the benefits of the community was having access to a podcast analytics/insights service that provides the data needed for the aforementioned opportunities. Which meant I needed a hosting provider that supported a third-party tracking prefix.

This is when I learned that Anchor doesn’t support third-party prefixes.

The Anchor community has made requests for it, but they were pretty much met with a flat ‘no’ or silence. I found no other free platforms that supported prefixes, and one detailed post by Acast about why they specifically do not.

There goes the free criterion.

Video podcasting support

Video podcasting is pretty new and support by podcatchers/players is a bit mixed, though the major ones generally do it. Being a person in tech, I tend to view bleeding-edge things with a certain level of wariness, but STEAM Powered has been released in video format on YouTube since day one so it seemed logical to look for a platform that supported it.

Long story short, there aren’t many, and the only one that was affordable on my self-funded budget was Podbean. The migration from Anchor was pretty seamless, and the UI, while clumsy, was serviceable, so I paid a year in advance to upgrade to the tier that supports video and thought that was that.

Narrator: That was, in fact, not that.

As I mentioned before, Anchor supports video podcasts but it was a relatively new feature at the time and as I knew I’d be migrating away and couldn’t guarantee the new platform would support video, I didn’t bother using it. If I had, I may have discovered my problems with Podbean sooner.

![Screenshot of Podbean pricing table showing Unlimited storage and bandwidth](../attachments/podbean-pricing.png "Excerpt from Podbean’s pricing table highlighting their storage entitlements"]]

Podbean’s pricing page shows their paid plans as having Unlimited storage and unmetered bandwidth subject to their Fair Usage Policy, which isn’t linked but mentioned in their Podcaster Terms. What I didn’t realise was that there was a file size limit, which isn’t indicated in the marketing pages or UI until you try to upload your file and it errors.

Sadly, it didn’t occur to me to check if there was a limit in their support documentation after I saw the word ‘unlimited’. So that’s on me. You can find the file format and file size information here.

“But the limit is 3GB!” I hear you say. “That’s heaps!”

3GB is heaps, but when you record in 1080p and your episodes average more than an hour and can run for 90 minutes, 3GB isn’t as ‘heaps’ as one would like. It took a while, but with a little direction from Podbean support and help from Arielle Nissenblatt’s wonderful Podcasting Community Discord, I managed to get my longer episode just a hair under 3GB by exporting in 720p with a few additional settings tweaks. Annoying that I have to drop the res, but whatever works, right?

Narrator: Except when it doesn’t work.

This is when I learned that video podcasts with Podbean won’t be picked up because as a “passthrough Partner”, Spotify will only accept mp3 from them. They will not convert mp4 to mp3, and there isn’t a way to provide a feed that provides both formats (that Spotify will read). Spotify is also about 30% of my listeners.

It is at this time that I realised that I had paid for a service that was not fit for purpose. However, I could not cancel without losing what I paid because Podbean has a no-refund policy and does not prorate. I also couldn’t find a way to downgrade my plan, and even if I did that would mean I would be stuck with Podbean even longer than the year I had pre-paid. And so, I wait.

Podbean is not for me. However, it might be for you if you:

  • have a video podcast you don’t want to publish to Spotify, or
  • do not have a high-res long-form video podcast and can fit your episodes within 3GB, or
  • you have an audio-only podcast

A couple of other things to be aware of which I have put down to ‘creative differences’:3

  • filenames of media you upload must be unique, ie. no two images called cover.jpg even if it’s for different episodes, otherwise the other episode(s) that use that filename will all use that one image (I had a couple of episodes all showing the same image for a while until a listener pointed it out to me)
  • replacing the audio/video media for an episode with a file with a different extension will cause your stats for that episode to be lost, just use mp3 and stick to it

In case it’s not clear, I’m moving away from Podbean. They are not suited to my needs, and as I am starting a second podcast soon, I will be moving to a platform that will allow me to run multiple podcasts sustainably. I’ll let you know how that goes once I’ve settled in.

I’ve also dropped the video podcast support requirement for my hosting platforms. After the problems I had with Podbean, I’ve decided it’s too bleeding-edge for me. I’ll keep publishing video on YouTube, but will stick to audio in the podcast platform space until it becomes more common as an offering.

Honestly, if it wasn’t for the need for third-party prefix support, I wouldn’t have left Anchor. But at the time I was a year and a half into producing a niche podcast with a small audience and as much as I love it, it’s quite frankly a time and money sink. I felt I needed to get serious about metrics and quantifying what I was doing in order to improve, grow, and at the very least, cover my costs.

Third-party tracking prefixes revisited, and analytics

I want to circle back to third-party tracking prefixes.

Most platforms provide some sort of analytics or stats reporting for your podcast. However, while we’re at a stage of the internet where services like to tell you that you own your content and are free to move it wherever you please, and provide the tools to import and export your data at will, no such thing really exists for podcast statistics.

At this time, there don’t appear to be any standards for what a single record should look like, so even if you are able to export your data, you can’t really move it anywhere else. Not to mention what you get in the exports can be pretty skint and not something you could plug into your own reporting tools, let alone anyone else's, and get much value out of.

It’s one reason why third-party services (and their prefixes) are useful but you’re effectively transferring this data lock-in problem from your hosting platform to an analytics service. If you do have to switch third-party analytics services, your data is just as gone.

I told you how I feel about bleeding-edge, but I’m going to make an exception for the relatively new Open Podcast Prefix Project (OP3) because if it allows people to free their data from these services, I’m rather for it.

The Open Podcast Prefix Project (OP3) is a free and open-source podcast prefix analytics service committed to open data and listener privacy.

…

Once you start using this prefix, you are opening up your show's listener numbers to the world, without compromising your listeners privacy. Developers and companies can start building products and services for you with the data, without you having to manage a custom integration.

OP3 is open-source. The code is available on GitHub for all to see. Your podcast statistics, if you add their prefix to your feeds, are also available for all to see and available via the OP3 API.

If you don’t want your stats to be transparent, don’t use OP3, but platforms and services like Transistor, Castopod, Podlove, podCloud (fr), and likely soon more are getting onboard.

Even though there aren’t any reporting interfaces for OP3 yet, and my hosting platform doesn’t support it, I’m stacking the OP3 prefix on top of the third-party prefix I’m required to use for OSSA so that I can ensure my data isn’t locked-in, and if someone else doesn’t come up with something first, I am free to try my hand at doing reporting on my own data.

What do I know now?

Well, I certainly check what file size limits a platform has. Even though I’m not worrying about video podcast distribution anymore, if a platform specifies limitations, and they decide to add support for video down the track, I might be disadvantaged as a long-form video podcast. It isn’t a show-stopper, but it’s something to consider.

Export all the data you can from your podcasting or third-party stats hosting services when you migrate away. If they offer it up in different formats or views, grab them all because at the very least you can do some simple things in a spreadsheet when you need the info.

If stats are important to you, start using OP3 sooner than later. The data will be collected independently of whatever your services are doing, and it’s only a matter of time before someone starts a service or releases a project on GitHub that will allow you to query your data more easily.

It’s a little too early (for me) to adopt video podcasting through the existing ecosystem. But don’t let that stop you from recording or otherwise producing video content for your podcast. I am fully advocating short-form video content as a promotional and community-building tool for podcasters, and if you can, release your podcasts on video platforms even if they’re still audio-only.

I recently did the YouTube News Creator Workshop, and even though YouTube Podcasts is currently only in beta in the US, the facilitators encouraged users to publish their podcasts on YouTube, even if the video is just a static image for the duration of the video. This isn’t a new thing either, Tim Ferriss was doing this years ago for the Tim Ferriss podcast. Now, we’re not all Tim Ferriss, but it’s a pretty low-barrier thing to do to add your content to another platform and YouTube wants to encourage creators to explore different content formats on their platform.

When it comes to choosing a service, ‘you don’t know what you don’t know’ but you can do your best. Do your due diligence and read through the features and limitations of the service both on their marketing pages as well as in online communities and reviews. If you know your requirements are specific, check these things in the relevant knowledgebases and contact support to ask questions about your use case. I thought I did quite well with this, but I was still caught out by my own assumptions.

So, don’t be afraid to pay month-to-month until you’re certain the service does everything that you need the way you want to do it. Poking around during a trial period isn’t always enough, and you’re better off being able to test all the features you need to use as part of your regular workflow, and know you are able to do it seamlessly a couple of times before you stump up for the year. It might cost you a few extra dollars, but that’s not bad for peace of mind.

Have any hard (or easy) lessons to share? Have feelings about bleeding-edge? Let me know.

Stay curious,

— Michele

Cover image modified from Lexica.

Footnotes

  1. Setting aside feelings about its parent company Spotify. ↩

  2. I’ll likely write about communities at a later date, but I was looking for an active podcasting community with some common interests for support and sharing knowledge. The monetisation side (another thing I’ll write about later) was a nice-to-have. ↩

  3. I call them bugs, Podbean calls them features. ↩

Published January 15, 2023