The world of podcast metrics can get complicated. We have downloads, streams, performs, viewers, listeners, views, watchers, group members, and possibly much more phrases than that. However if you boil all of it down, I believe it comes to 3 primary labels: downloads, performs, and viewers. Here is what every one really means, and the caveats you’ll want to know for every.
1. Downloads (and “streams”)
That is the basic measurement. For a few years, the benchmark query was merely, “What number of downloads are you getting?”
A “stream” is normally nonetheless a obtain. In most trendy podcast apps, you possibly can open an episode, press play, and begin listening inside a number of seconds. That looks like streaming, however what’s actually occurring is a progressive obtain. The file begins downloading the second you press play and buffers forward within the background. As a result of connections are quick, it seems instantaneous. Some apps or knowledge networks obtain in chunks (50 MB, 100 MB, and so forth), so an extended episode may obtain midway first and end the remaining if you attain that spot. Different apps obtain the entire thing directly. Both method, the file is being downloaded from the web onto your gadget. That is nonetheless a obtain.
Downloads are essentially the most elementary measurement of a podcast’s attain as a result of they’re tracked on the server degree. However they’re dependable, not correct. There are many bots on the market, each good and unhealthy, that obtain episodes to transcribe, cache, or collect data routinely with out a human ever urgent play. Internet hosting suppliers like Captivate, Buzzsprout, Blubrry, Libsyn, and Transistor keep lists of what they take into account bots and filter these out of your counts (whereas normally nonetheless serving the file so nothing breaks).
Even with excellent filtering, downloads do not inform the entire story. I’ve talked earlier than in regards to the fable of month-to-month downloads. Saying “we acquired 100,000 downloads this month” tells you virtually nothing about your viewers dimension. Here is an excessive instance for instance the purpose: you can launch 100,000 episodes all downloaded by one individual, and your viewers can be one. So downloads per day, week, or month is basically a meaningless metric for understanding viewers.
The one place downloads-per-month is beneficial is promoting. When you have dynamically or programmatically inserted adverts throughout your entire catalog, downloads per interval tells you what number of advert placements may very well be served.
However if you wish to perceive your viewers dimension, the metric to make use of is downloads per episode after 30 days. That is the basic benchmark used for promoting and for understanding your viewers, as a result of it captures each your most loyal listeners (who obtain within the first few days) and the stragglers. Simply bear in mind it solely applies to an episode that is really 30 days previous, and it’s best to examine episodes utilizing the identical period of time since publication. Many analytics instruments (OP3, Captivate, Transistor, Blubrry, Buzzsprout) now present numerous benchmark intervals like 1, 3, 7, 10, 15, and 30 days, however all of them embody 30 days as a result of that is the long-standing benchmark.
This strategy additionally handles the messiness of IP addresses. If you happen to attempt to rely “distinctive listeners” by counting distinctive IP addresses in a month, you will get a quantity that is too massive. One individual can rack up a number of IP addresses for a single episode: they begin streaming at dwelling on Wi-Fi, get within the automotive the place the cell IP modifications, then pull by a espresso store drive-through the place the cellphone briefly hops onto that Wi-Fi. That is probably three IP addresses for one episode. Throughout a number of episodes, one individual may signify dwelling, work, and cell addresses. You’ll be able to strive dividing by the variety of episodes launched, however that usually cuts the quantity down too aggressively.
Downloads per episode after 30 days sidesteps quite a lot of that, as a result of as soon as an individual downloads an episode they in all probability will not obtain it once more. And when an IP handle modifications mid-download, many internet hosting and analytics suppliers are sensible sufficient to see the identical consumer agent selecting up proper the place the final chunk stopped and sew it again collectively as one individual.
The takeaway: downloads are very dependable however not very correct. They require quite a lot of filtering and algorithms to account for each state of affairs, which is strictly why we want different methods to measure attain.
2. Performs
So what’s a play? That is possibly the query Shakespeare would ask if he have been a podcaster in the present day. To play or to not play?
In locations like Apple Podcasts, a play is counted when the episode is performed. That doesn’t imply it was listened to fully or completed. It simply means it was performed. That is why you may have a look at your stats and see your performs are literally a lot increased than your downloads or your follower rely. It counts each play. If somebody pauses and presses play once more, that is one other play. If a voice assistant pauses playback to present driving instructions after which resumes, that may be one other play. Drive throughout city with turn-by-turn instructions interrupting your episode and also you may single-handedly account for dozens of performs.
So performs are attention-grabbing however not very dependable. That is the place the Alliance for Measurement in Podcasting (AMP) is available in. They seem to be a group that is been working behind the scenes, like a person behind the scenes, making an attempt to outline what a play really is. I do not love how secretive they’re being or who they’re leaving out of the group, however I do like what they’re making an attempt to do. If we may all agree on what counts as a play, that will be genuinely useful, particularly if it may lengthen to non-podcast platforms like YouTube.
For comparability: on YouTube it is assumed (by no means confirmed) {that a} view is not counted till a minimum of 30 seconds have performed. In podcasting, the IAB (Interactive Promoting Bureau) obtain tips say a obtain counts as soon as 60 seconds of precise audio has been downloaded, measured after header data like chapters, ID3 metadata, and the embedded picture. AMP is making an attempt to outline a play as 30 seconds, they usually may outline whether or not resuming after a pause counts as a brand new play.
The core problem is the place every factor is measured. Downloads occur on the server, so internet hosting suppliers and third-party analytics can see them. Performs occur contained in the app, the place these providers cannot see what is going on on. That is additionally work that would match the Podcast Requirements Challenge (PSP), and it is why I’ve lengthy wished an ordinary method of measuring issues. After we use a time period, the vast majority of platforms and apps ought to perceive it to imply primarily the identical factor. The IAB obtain tips tried to try this for downloads, however there’s nonetheless wiggle room and methods it may be faked.
Particular thanks
- Ralph Estep Jr. invited me onto Podcasting Morning Present to speak about podcast engagement, chapters, and extra podcasting subjects.
- Martin Lindeskog streamed 706§ and boosted 1,111§ saying, “I broke your ‘rule’ with my first podcast, EGO NetCast. It’s caricature of yours actually, created by editorial cartoonist and tremendous artist, John Cox. I’m carrying a baseball cap, holding a retro microphone and a radio. Considerate recommendation about not carrying headphones. I’ve discovered a pair of very comfy headphones, AKG K92.”
3. Viewers
Some folks name this “listeners,” however I favor viewers. Podcasting has at all times included video, so “listeners” implies audio solely. “Viewers” covers each your listening viewers and your viewing viewers.
That is the third main solution to measure a podcast’s attain: what number of precise individuals are consuming your present. And sadly, there isn’t any common solution to measure it in all places. There are actual privateness issues. If a podcast app can observe what you take heed to through some ID tied to you, it may construct an image of your consumption and promote that to advertisers. Some app builders have flatly refused to do something that identifies an individual or reviews the variety of customers, and viewers needs to be tracked contained in the app to start with.
Regardless of the privateness issues, there are methods this may very well be achieved. One thought, mentioned by Dave Jones behind Podcasting 2.0, is the Common Listener ID or Common Viewers ID (ULID/UAID). The idea: a podcast app generates a random string of letters and numbers assigned to an individual and sends it with every obtain request. It doesn’t matter what IP handle that individual is on, their downloads get tracked underneath that one ID.
However this raises privateness issues of its personal. If an app makes use of the identical ID throughout a number of podcasts, analytics instruments may begin seeing every thing you take heed to. Apps may randomize it so that you get a novel, globally distinctive ID per podcast. But even then, a malicious social gathering watching the server may observe all of the IP addresses tied to at least one ID and construct an image of the place you go. I would not try this, and most internet hosting suppliers would not both, however advertisers would like it.
This is similar actuality behind “my cellphone is listening to me.” The reality is scarier: your cellphone is not listening, the algorithm is watching. It notices that you just and one other individual continuously go to the identical place across the identical time and share social connections, so it infers you in all probability discuss. That sort of profiling is way extra invasive than eavesdropping, and it really works by constructing a profile of somebody, usually with out significant consent. On social apps you technically consent by phrases of service that everybody simply clicks “agree” on. Podcast apps do not need to require that sort of consent, although Apple Podcasts does generate a novel ID for you (one you possibly can reset anytime in your iPhone or Mac, although virtually nobody does).
There’s additionally the handoff threat seen with browser extensions and plugins: you agree to at least one privateness coverage, the developer sells the challenge, and the brand new proprietor modifications what’s tracked. It is occurred with Chrome extensions, Firefox extensions, and WordPress plugins. A ULID/UAID may completely be applied to respect privateness, however provided that each app developer commits to it and people commitments maintain.
Crucial metric is the one which measures your objectives
The explanation I made this episode is that can assist you perceive the caveats. I am a little bit of a Mr. Caveat. However when you perceive how downloads, performs, and viewers are every measured, you possibly can learn any platform’s numbers with the correct context as an alternative of chasing the most recent metric or commonplace.
For instance, in Apple Podcasts you possibly can see detailed consumption knowledge as a result of they observe it inside their ecosystem. Realizing that performs counts each play, you will perceive why an extended episode racks up extra performs (folks pause and resume lengthy episodes extra), and {that a} increased play rely doesn’t imply a much bigger viewers. Performs doesn’t equal viewers. Downloads is nearer to equaling viewers, however even a obtain would not assure somebody really consumed the episode. Edison Analysis has shared survey knowledge suggesting most individuals who obtain an episode take heed to most or all of it, however which may be shifting as extra “downloads” are actually progressive streams, and plenty of listeners do not know the distinction.
So it actually comes again to what issues for you. A few of these numbers genuinely matter when you’ve got sponsors, advertisers, or a enterprise that wants KPIs. There’s an actual want for that in these circumstances. However for a lot of podcasters, the gas to maintain going is not the dimensions of the viewers, it is the depth of your relationship with the viewers. The extra you foster that relationship, the extra that turns into the quantity that issues.
How many individuals emailed you this week? What number of commented in your episode? And this is one thing cool I present inside Podgagement: not simply your complete opinions, however the ratio of rankings to opinions. In Apple Podcasts you possibly can go away a score with out a evaluate, however each evaluate features a score. A smaller hole between rankings and opinions alerts a extra engaged viewers. For The Audacity to Podcast, that hole is way smaller than for a present like Serial, which has an enormous variety of rankings however a giant hole earlier than opinions. If podcast opinions matter to you, accumulate and observe them with Podgagement!
After all, even opinions are one per individual, so possibly the metric that issues most for you is suggestions, group engagement, or social reposts. Take into consideration why you are doing all of your podcast within the first place. Take into consideration your P.R.O.F.I.T.: reputation, relationships, alternatives, enjoyable, earnings, or tangibles. Which metric measures the why to your present? Is it viewers dimension, performs, downloads, or one thing else totally? If enjoyable is your aim, then possibly the one metric you want is an easy thumbs up after every episode: did we’ve enjoyable? If sure, you met your aim, and it would not matter what number of downloads, performs, or viewers you had.
It is as much as you to determine what’s vital to your present. Crucial metric is at all times the one which measures your objectives.
If you happen to love The Audacity to Podcast and worth the podcasting inspiration and schooling I present, would you please take into account giving again what it is price to you?
Supercharge your podcast engagement and develop your podcast!
Do you ever really feel like your podcast is caught? Such as you’re pouring your coronary heart into your podcast nevertheless it looks as if nobody is listening?
Attempt Podgagement that can assist you supercharge your podcast endgagement!
Get speakable pages to simplify partaking along with your viewers, settle for voicemail suggestions (with automated transcripts), see and share your rankings and opinions from practically 200 locations, observe your podcast rankings throughout practically 34,000 world charts, uncover networking alternatives, and extra!
Ask your questions or share your suggestions
Comply with The Audacity to Podcast
Disclosure
This publish might comprise hyperlinks to services or products with which I’ve an affiliate relationship. I could obtain compensation out of your actions by such hyperlinks. Nonetheless, I do not let that corrupt my perspective and I do not advocate solely associates.
