Reading time: 15 minutes
Knowing your way around podcast analytics is absolutely essential if you want to grow your audience. As the saying goes, you can’t improve what you can’t measure. Yet podcast analytics can be hard to navigate. As a podcast publisher, you’ve probably asked yourself at least one of these questions:
- Which dashboard should I use? Apple Podcasts Connect, Spotify for Podcasters, or that of my hosting provider?
- What is the difference between a download, a play, a stream, a start, a listener, and a follower? Why are there so many of these metrics and why do they all sound vaguely similar?
- Are all metrics in my dashboard equally important? Which ones should I focus on to measure the success of my show and know what content to double down on?
If this sounds like you, this article is going to bring you a lot of clarity! By the end of it, you will know:
- Which dashboard you should be using
- The differences between all podcast metrics
- Which metrics will help you understand how to make your show more compelling for listeners
- Which metrics help platforms decide whether or not to recommend your content
- Which metrics will help you convince advertisers to sponsor your show
- Which metrics are a waste of your time
- Other frequently asked questions
Which dashboard should I use?
Podcast publishers can get data about their show from 3 different sources:
- The analytics dashboard that comes with their hosting plan
- Spotify data can be found on Spotify for podcasters
- Apple data can be found on Apple Podcasts Connect
So which one should you use? Well, all of them, because each of them provides data that the other 2 can’t report on.
- Your hosting provider reports all downloads, no matter the platform or the app, which means it is the best suited to give you a bird’s-eye view of your total number of downloads. This is essential because it determines the monetization potential of your show.
- Contrary to your hosting provider, Spotify and Apple only report on the listening activity in their own app, but the level of detail they can provide is unmatched.
In the next section, I will show you exactly which platform you should use to monitor which metric.
The differences between all podcast metrics
All platforms mentioned in the first section tend to use similar-looking metrics but they turn out to be quite different once you look into their definitions. This section is here to help you understand how they differ and more importantly, which are flawed/outdated and which ones you should focus on.
Downloads
Usefulness: 4/5
Platform: your hosting provider
As the name lets on, a download happens when an episode’s audio file is stored on a device (smartphone, tablet, etc.). Downloads are still the most widely used unit in the podcast advertising industry with advertisers booking podcast ad inventory on a CPM (cost per 1000 downloads) base.
Podcast download benchmarks
Many podcast publishers want to know how they compare with other shows and how many downloads per episode is a good number. First, we’re going to show you a benchmark because we know you don’t want to read a wall of introductory text.
Here’s the distribution within 7 days of releasing an episode:
And here’s the distribution of downloads irrespective of time since publication:
Now that you’ve seen the numbers, we would like to add an important note:
This data is interesting but take it with a grain of salt because it lacks context. If you really wanted to use benchmarks to measure your progression, you’d need to compare yourself with podcasts in the same niche because each genre and even sub-niche has its own natural audience ceiling.
Additionally, you’d need to know when they released their first episode, how many episodes they’ve released to date, and if the host already had an audience prior to starting their podcast that they could leverage to jump-start their audience growth.
Say you had a show in the true crime niche and you launched it 6 months ago. If you wanted to compare it to a show like Crime Junkie, you’d need to know how many downloads they were generating 6 months post-launch.
Unfortunately, there’s not a lot of data available to be able to make apt comparisons so we advise you to focus on what you actually have control on: the pace of your audience growth.
How to look at your download data?
Usually, when you log into the analytics dashboard of your hosting providers you get something like this:
Or like that:
The problem is that these numbers aren’t that useful.
With a few exceptions, most dashboards provided by hosting providers are not well suited to perform a proper analysis of your download data.
A decent dashboard should give you the ability to freely choose
- the time scale (daily, weekly, monthly)
- the start and end date of the period you want to analyze
This makes it much easier to:
- focus on trends rather than the noise of recent daily data
- see where trends start and reverse so you can better explain them
- spot outliers (did you get featured?)
- match the data with complementary information (for instance if you were to look at weekly data and see a sharp decline in calendar week 39, you would be more likely to remember and map it to the fact that you were sick that week and didn’t publish).
Your best option is to manually extract these numbers at the end of each week and month into a Google sheet and build yourself a rudimentary dashboard.
To close the topic of downloads, I want to add that downloads probably will lose their status as currency of podcast advertising in the coming years. Technically speaking, a download doesn’t necessarily equate to a listen because many podcast apps download new episodes by default (Apple only does this if you follow a show). As the industry matures, more and more advertisers will realize this and probably push for a change to make sure their ad dollars are spent on actual listens. This is the reason why didn’t give downloads a 5/5.
Plays
Usefulness: 1/5
Platform: Apple
The total number of times people pressed play on your episode (as reported by Apple). This metric is not too reliable either. As you can see in the screenshot below, the staggering difference between listeners and plays can only be explained in one way: a play is literally the action of pressing the ‘play’ button. If you paused and pressed play again, that’s an additional play.
Image courtesy of Apple
Starts:
Usefulness: 1/5
Platform: Spotify
According to Spotify, “starts are recorded every time a listener starts one of your episodes”.
Streams:
Usefulness: 2/5
Platform: Spotify
Streams are more meaningful than plays and starts because per Spotify’s definition, a stream is only counted after someone has listened to at least 60 seconds of an episode. There is an important caveat: this metric is only used by Spotify for now which means you cannot compare apples to apples (pun intended) with other platforms.
Listeners
Usefulness: 5/5
Platform: Apple & Spotify
The total number of people who have listened to your show. I like this metric a lot because both Apple Podcasts and Spotify use it and they define it the same way. Both platforms count unique listeners which has the benefit of giving you the exact size of your audience at the show and episode level.
It might be tempting to get your number of unique listeners from the podcast analytics dashboard provided by your hosting provider but I recommend getting this data directly from Spotify for podcasters and Apple Podcasts Connect. Without getting too technical, Spotify and Apple rely on first-party data which podcast hosting providers don’t have access to.
Because consumption is fragmented across many platforms/apps, you need to use a quick workaround to get a close approximation of your total count of unique listeners across platforms. Here are step-by-step instructions:
- Log into your Spotify for Podcasters and Apple Podcasts Connect accounts.
- Add up the number of listeners you get on both platforms (a)
- Go to your podcast hosting analytics and look up the consumption by platform.
- Add up the % for Apple and Spotify (b)
- Calculate a/b. For example, if you have a total of 1000 listeners on Spotify and Apple Podcasts combined and these platforms account for 60% of podcast consumption, you have approximately 1000/60% = 1667 listeners
Followers
Usefulness: 5/5
Platform: Apple & Spotify
Followers are people who have clicked on the follow button, which is the equivalent of the subscribe button on Youtube. A follow is great for 2 reasons:
- It is a trust signal for platforms showing that a user wants to hear more of your show which is probably used to determine show recommendations.
- In Apple Podcasts, a follower will get notified by default when a new episode comes out. This increases the odds that they will listen to more episodes in the future which in turn means more revenue for you.
Note: Spotify users can follow without getting notified automatically, they need to press the bell icon to opt in, similar to Youtube.
By knowing your follower count you can derive one very useful metric: your number of downloads per follower (formula: #downloads / #followers). It will tell you more about the consumption behavior of your followers. Do they stick around or follow and forget?
Which metrics can help you understand how to make your show more compelling for listeners
Streams/starts
Usefulness: 4/5
Platform: Spotify
Formula: #streams / #starts
This ratio gives you the percentage of listeners who make it to the 60-second mark. It is not provided as is on the Spotify platform but you can manually calculate it by dividing your number of streams by your number of starts.
This will give you a number between 0 and 1. You want this number to be as close to 1 as possible. If you notice that your stream/start ratio is too low, you need to change the intro of your episodes by including an extract that contains the most fun or interesting piece of information right at the beginning. Think of a movie trailer. Most of the good bits are included to get people to go watch the movie.
Completion rate
Usefulness: 4/5
Platform: Spotify
In the Spotify for podcaster dashboard, navigate to the episode performance tab to see. You’ll see a graph that represents your audience’s retention. It includes the average time users spent listening to the episode, plus the percentage of listeners who listened to 25%, 50%, 75%, and 100% of your episode. This is helpful to see where people dropped off and identify ‘weak’ portions in your episode that you should work on avoiding in the next ones.
Average Consumption
Usefulness: 4/5
Platform: Apple
In Apple Podcasts Connect, navigate to the ‘episodes’ tab and you will see a table showing you the performance of all of your episodes. The right-most column shows you the average consumption in %. It shows how much of each episode users have listened to on average. Comparing the performance of each show helps you determine which of the topics you’ve covered so far did the best job at keeping your listeners engaged.
Engaged listeners
Usefulness: 4/5
Platform: Apple
This critical data point tells you the % of listeners who’ve listened to at least 20 minutes or 40% of an episode. This is great to gauge the stickiness of your content. Though Apple has released no official statement, it is not too far-fetched to assume that Apple Podcast’s algorithm uses the ratio of engaged listeners (engaged listeners/listeners) as a quality signal to determine show recommendations.
Follower/listener ratio:
Usefulness: 5/5
Platform: manual calculation based on Apple and Spotify data
Formula: #followers / #listeners
If it’s trending upward, it means that your latest content has helped to turn more listeners into loyal followers. This ratio is crucial if you are spending money to acquire listeners (by running ads on Facebook or TikTok for example). If the ratio goes down too much then it means you are either not targeting the right listeners or you are promoting the wrong episodes.
Which metrics help platforms decide whether or not to recommend your content
This section is going to be short because it’s pretty much a copy-and-paste of the previous section. If you think about it, it makes a lot of sense.
Platforms like Apple and Spotify want users to use their apps as often as possible to maximize retention and monetization. And how do they make sure users stick around? By providing the best experience possible, which means recommending the most relevant and entertaining shows and episodes.
They assess this by looking at ratios like:
- followers/listeners (in %)
- engaged listeners/listeners (in %)
- completion rate
2 additional metrics not mentioned in the previous section that Apple looks at are:
- Your average rating on Apple Podcasts (note that Spotify is progressively releasing this feature on their platform as well).
- Your number of reviews on Apple Podcasts. Depending on the geo, a very limited number of reviews can move the needle a lot. We have seen shows improve their rankings in Australia after receiving half a dozen reviews.
Which metrics will help you convince advertisers to sponsor your show
On top of downloads which we’ve talked about at length earlier in this article (click here if you need a refresher), there are 2 other major data points advertisers want to know about before they decide to run ads on your podcast:
Gender and age distribution
The best place to get this data is Spotify. They can rely on self-reported reported data users provide during the sign-up process as well as data from Facebook if users choose to use Facebook Connect.
As a podcast publisher, you should be aware that Spotify as a platform skews younger so the data is not 100% representative of your entire show’s audience. That being said, it is the most accurate data you can get, most advertisers don’t even realize the data is slightly biased and will happily take it.
Additionally, these numbers can inform the choice of social media platforms you choose to promote your podcast on. For instance, if you find out that your audience is mostly made up of gen Z males then you probably want to invest your efforts in YT shorts and TikTok rather than say, Facebook and Instagram.
Geographic distribution (country level):
As an advertiser, I want to reduce wastage as much as possible and spend my budget on listeners that can actually purchase my product. For example, If I only deliver in the US, I will avoid advertising on podcasts whose audience is mainly located in Europe.
Which metrics are ‘vanity’ or even downright useless metrics
More data doesn’t necessarily mean better data. Some hosting providers try to compete on thoroughness rather than usefulness. Podcast publishers end up drowning in a sea of useless data that has the appearance of depth and looks important but it isn’t really actionable.
Here are data points that you can disregard:
Listening by city
Yes, it is fun to look at and share with your podcast friends, but there’s not much you can do about it. Local businesses would only care about it if the overwhelming majority of your audience was located in their area which is very rarely the case.
Listening by device
Same as the above. Hosting providers have access to this data so they include it in their dashboards to appear more professional and thorough but what’s the point? You cannot optimize your content for different devices.
What podcast listening app do people use to listen to my show
Again, not much you can do about it or learn to inform your content creation plan. Ok, I might be a bit too harsh with this one, there actually is one case where keeping an eye on it actually makes sense:
If an app with a smaller market share suddenly gets an abnormally high count, then you probably have been featured in that app and you don’t know it yet. It would be a good idea to reach out to the team behind the app and start building rapport as it might bring more features in the future.
Frequently Asked Questions
What is AppleCoreMedia in the dashboard of my analytics provider?
AppleCoreMedia downloads are a mixed bag of downloads happening in the Apple Podcasts app and in other apps on Apple mobile devices. When a user starts listening to an episode without having downloaded it beforehand, the user agent Apple sends to hosting providers changes from the name of the actual app to AppleCoreMedia.
Podnews has published a very well-researched article on the topic which you can find here. The long and the short of it is that they found out 42% of their AppleCoreMedia traffic verifiably came from apps that are not the Apple Podcasts app (e.g PocketCasts, Overcast, etc)
How can I easily see if one of my old episodes went viral?
The easiest way I have found is to look at the episode performance view in Apple Podcasts Connect. Normally, the consumption of an episode over time follows a similar pattern which takes the form of an asymptotic curve as shown below. If one of your old episodes goes viral, the graph representing the listeners over time will show a sudden sharp increase at a point in time where the curve should normally get flatter and flatter.
Image courtesy of Apple