AfterPod — Different Statistics
Here’s some actual information regarding the podcasting space. It is gleaned from a dataset that, though not complete, is fairly current and large enough to be representative. This discussion looks more at the HOSTING side of the podcast endeavor. When WE listen to a podcast, whether downloaded or streamed, where does it come from?
Sometimes, extra knowledge can be obtained with a very small investment… in this case, the time to query a database. I’ll note that podcast hosting is NOT a goal of my primary project, but I became curious while reviewing a related group of 250,000+ database records I am cultivating in support of my web media application.
Here’s what I found and one reaction when I shared it.
Who “serves” the RSS feeds?
There’s been a lot more analysis of podcasting lately, most focused on in-episode reporting, how we behave while listening to an episode (how long do we listening? are parts skipped?).
Another view could be based on the volume of downloads/listens — indicating the popularity — of podcasts. (How many people are listening?)
Yet another look might rank them by hosting income or advertising revenue.
The view above has NOTHING TO DO with any of that. The list merely extracts the main domain name from the RSS feed and keeps count of the times it appears in the record list. It does not consider possible redirects or other such networking maneuvers that can and do occur.
A few things to note about some of the larger groups:
- Feedburner (a Google service) does not actually host the feeds, it just facilitates their delivery. Many such “middlemen” tend to anonymize the identity of the actual feed source as a by-product.
- A couple of categories, NULL and UNKNOWN, are… well… the “known unknowns” you hear about. As the data is refined, these will go away. Or into other categories. I hope.
- Apple.com? It appears they aggregate a lot of .EDU seminars and presentations. Many are video.
- Librivox and some others are not really podcasts, they are audiobooks (usually public domain) that are read and recorded by a worldwide network of volunteers. Do I need to say FREE?
Inside B̶a̶s̶e̶b̶a̶l̶l̶ Podcasting
So, I shared the information within a couple of podcast circles… represented it as a “glance” at the podcast hosting space. While general information is known about this market, specifics are harder to come by. I like putting pieces of the puzzle next to each other.
SOME, however, became defensive because it didn’t give the FULL picture. Blubrry, for one, doesn’t rank high on the list and took umbrage as they are a “bigger player” in the industry (which everyone in the industry already knows).
A perfect example of why data requires interpretation and refinement.
The list examined the SOURCE OF THE RSS FEED, which just gives your podcast player info to find the actual podcast media file on the internet. The location of that file may or may not be the same domain as the podcast’s RSS feed.
Blubrry mostly provides a WordPress-based solution in the podcast hosting arena and, as such, most of their customers operate primarily under their own domain name (which is a smart thing to do, BTW). If you look at the graph at the top of the list above, it shows only the top 100 domains because they aggregate podcast RSS distribution… there are thousands more that operate as single entities.
Todd at Blubrry felt a little dissed by the data above, as real as it is.
So I did another query.
Remember, this is a moving target — live data — so things don’t always match up exactly, knowwhatImean?
Who “serves” the podcast media files?
Now, by looking not at where the RSS feed comes from but rather at where the MEDIA is sourced, we see a similar view but with some big differences.
Blubrry went from have a count of 705 in the first group to over 11,000. This is relative positioning closer to what those-in-the-know expect.
It’s also worth mentioning that, with the recent melding of BlogTalkRadio and Spreaker (and Voxnest!), those two can be combined into one, larger entity.
Oh, podtrac.com is not really hosting, they gather stats and do analytics. Another middleman… here’s what that group looks like:
FWIW.
EDIT: Extended on this in another article about Dormant vs Dead.
TECH TYPES:
If you are interested in emerging web media development projects, check out my VIZdex Annotated Timeline Presentation Platform and its first deployment,podDVR.com.
Also, here’s my last article about paying attention while walking around.
You may want to read the first article, which establishes much of the foundation for these views, or the second one that begins to forth our intent to purposefully (and positively) modify podcast consumption behavior.
