Podcast Stats – Downloads, Listeners, Subscribers
This is the second post in my series focusing on the stats that are important to us as podcasters. See the first post for the complete table of contents.
If you're like me, you use Libsyn to host your podcast media files and your RSS feed.
I love libsyn because the stats make it easy to see how many times someone downloads my show (over 10,000 times!). Libsyn ticks the download counter every time someone streams or downloads an episode of Amazing Workplace. It is an great snapshot metric for the popularity of my show, but it doesn’t represent the total number of listeners I have. The download counter doesn’t care what happens after it serves the file. It doesn’t report if it was a person or a bot making the request, and doesn’t tell me if the file was ever played. One person could request an episode thousands of times (or a smart person could just write a script to do it). Someone could be switching devices in the middle of an episode, which counts as two downloads, not 1. Or they could be like me and pause a non-downloaded episode for days on end, and come back to finish it later (I love you Podcast Addict).
Downloads don’t equal listener counts. They are more closely related to the total number of times someone hears your show. Download numbers are an inaccurate representation if you're looking for specific numbers. Trends in the download numbers are great indicators of an audience’s growth or contraction.
Listeners Matter More Than Downloads
Another, more important metric for me, is the number of listeners I have. How do I find that information? Unfortunately, I have to rely on those pesky download numbers. In Libsyn, I can drill into a single episode and filter the downloads based on a date range. I look at a 4-week period, as that represents what the industry recognizes as the average lifespan of an episode (at least for those of you that may seek sponsors). Pick an episode, start at the release date and span the following month, and jot down the total downloads. Run this on several episodes that are at least one month old, and you'll get an idea of your number of listeners. Find the trend. Is that number of “listeners” climbing, or falling? This trend will inform your marketing decisions, and provides direct feedback to audience reach. It can also be the basis of negotiating sponsorships, but that's a subject for another article. Unfortunately, this information is not real-time, as it looks at a 4-week window.
Subscribers Matter More Than Listeners
If you're willing to play around with your release and marketing schedule, you can deduce subscriber counts from listener counts, . I release new episodes on Mondays, but I hold off on promoting them until Tuesday. I'm not advocating you do this. The only reason I do it is to “game” the stats systems and see how many downloads occur in a day with no promotions of a new episode. This gives me a feel for how many subscribers my show has.
Anyone that is a serious podcast listener is setup to download new episodes onto their device on release day. All I have to do is count the number of downloads on release day, and I have a decent measure of the number of subscribers. I call it a deduction because none of these are spot-on with accuracy. Listeners will be lower than 'downloads per episode.' Subscribers will be lower than what I'm guessing here - not everyone keeps to my schedule.
My deduced subscriber count sits about 25% of the estimated monthly listener number. Even though it is not a real measure of subscribers, it does inform my decisions. I’d like to double this number, and can make decisions off this goal. I need stronger calls to action on my site and in the podcast. Motivating people to subscribe and share the show should be one of my chief activities in my intro and outro.
So the 'equation' is:Downloads inform listeners, from which we can deduce subscribers.
All these stats are not accurate representations of the actual numbers. They are not a 'real' representation of episode plays, listeners, and subscribers. The trends that I observe and goals that I set and measure are real. The results are real, which is why I'm completely comfortable with knowing I'm looking at inflated numbers.
Of course, if anyone can figure out how to get real-time, accurate stats, give me a call.
A Note About Gaming The System.
Do it for good, not fraud.
It is possible for anyone to boost their download and listener numbers - go search “twitter bombing.” It is unethical, egotistical, and, if you work with sponsors, fraudulent. Advertisers want to reach a particular demographic, not just anyone. They want to see more than just the number of listeners you can guarantee them. They're interested in age, location, income, education, and on and on.
Depending on the contract that gets put in place, they'll also have the right to request an audit of your stats. If you've been inflating those, you may face an expensive lawsuit.Also, podcast hosts like blubrry are starting to find ways to remove these numbers from stats, so you’ll be a moving target that everyone is pointing at. Sooner or later, someone will score a hit and take you out, so just do things the right way and build your audience organically.
What is your metric?
You may not agree with everything I just shared, which is good - stats are only as good as the people that interpret them. We all have our own lens through which we observe the world, and we all have different tactics and priorities. What are your most important metrics? What stats are you checking over and over? What are you addicted to, and what insight does it add to your business?