How to market your podcast, part 5: Measuring success

Podcasting has found new favor with the marketing world. Marketers are creating podcasts left and right, but are spending so much time on creating it that they forget to market it. This series is for you, so that people listen or watch your new creation.


Posts in the How to Market Your Podcast series:

Interested in a real-life example? Check out my marketing podcast, Marketing Over Coffee!


In today’s installment, let’s talk about measuring the success of your podcast. It’s one of the most controversial and tricky areas of podcasting. In podcasting, there are 4 sets of metrics that you likely have access to: feed/file, reach, web, and marketing metrics. We’ll dive in from the top.

Feed/File Metrics

Feed and file download metrics are the earliest form of podcast metric. They are essentially the number of times your digital files have been downloaded and the number of people subscribed to your RSS feed. While basic in nature and easy to understand, they’re highly unreliable for two main reasons: caching and completion. Many services such as iTunes, Stitcher Radio, and other podcast listening software will make a copy of your episodes and serve them from their servers. This reduces load on your server, which is not a bad thing. However, one copy made of a file from your server can be played hundreds or thousands of times, and you won’t know.

Completion is the second reason why feed metrics are unreliable. Just because someone’s subscribed to your show or downloaded your episodes doesn’t mean they’ve listened to you. Here’s an example from the Freakonomics Podcast that’s on my phone:

IMG_4238.PNG

As you can see above, I’ve listened to none of these episodes yet. However, they still count as downloads. If you’re trying to gauge the success of your show, it’s relatively easy to see how misleading feed and file numbers can be.

Reach Metrics

Reach metrics are the top of the audience and media funnel:

blue_belt_slides_pptx.jpg
  • How many people in your audience have you exposed to your material?
  • How many have engaged?
  • How many have clicked on a podcast player in your Twitter feed, for example?

These numbers will come from a variety of sources like social media listening tools, Twitter and Facebook audience tools, and can be part of your overall podcast analytics package. If you’ve got a huge social following but no one clicks on anything, then you don’t have much reach into your audience.

Web Metrics

Downstream from audience and engagement metrics are things like web analytics. Assuming you’ve set up the basic content infrastructure for your podcast, you should have web analytics available to tell you how people are finding your show and how often they stick around.

GA podcast stats MOC.jpg

All of the basic web analytics metrics apply, such as new and returning visitors, form fills, and pageview/engagement times.

Business Metrics

Finally, down funnel even further are your business impact metrics. Who did something meaningful such as sign up for an email newsletter, ask to speak to a sales representative, or even buy something? Here’s an example of my on-site survey asking people how they heard about me, personally:

How_did_you_hear_about_my_site_.jpg

For this survey period, 2 out of the 13 people who filled out the survey indicated that my podcast was part of the way they found me. That’s fairly compelling.

The Big Picture

Each of these metrics should be lined up in a dashboard or report so that you can see – and your advertisers/sponsors can see – what kind of numbers you’ve got. I’ve taken metrics from each category and combined them into a spreadsheet, then charted it out for a big picture view of the health of the show:

Tableau_-_mocstats.jpg

What do we see?

  • [1] is the subscriber count. We’re looking for direction here more than anything, because of the aforementioned caching and completion problems.
  • [2] is the reach number. This combines media mentions, social shares, and other measurements.
  • [3] is website traffic, straight out of Google Analytics.
  • [4] is email list subscriber growth, the number of people signed up for the Marketing Over Coffee mailing list. Again, looking for trend here.

I could add in all kinds of other metrics, too, such as LinkedIn group members, Twitter followers, sponsor revenue, etc. in their appropriate buckets, but this is more an example than anything. When you do this for yourself, you will almost certainly have to resort to pulling all the data together in a spreadsheet. There are just too many data sources that aren’t connected to each other to build an all-in-one dashboard that isn’t custom-made.

Advanced Big Picture

Using the visualization and analysis techniques from my book Marketing Blue Belt, I can take the above data and figure out what the trends are:

Tableau_-_mocstats 2.jpg

The blue lines were simplified into 14 day moving averages for visual clarity. The black lines put atop them are trend lines (2 degree polynomial fitting) that show the general momentum of each data series. What does this tell me? The top two series, subscribers and reach, are going in the wrong direction. The bottom two lines, web analytics and email list growth, are doing okay. Thus, from this analysis, I should spend more time growing the audience in order to restore momentum there.

Wrapup

Each individual metric above has issues. Combined, they can paint a picture that gives us some sense of what to do and whether things are going in the right direction or not. There are more advanced podcast metrics available, but typically out of reach for smaller shows like mine. Tom Webster and his team at Edison Research have the scoop on the top shelf stuff.

This also concludes our how to market your podcast series. I’ll have either a book or webinar (or both) coming at some point in the future, but for now, this is a good start. Thanks for reading, and be sure to tune into Marketing Over Coffee!


Posts in the How to Market Your Podcast series:

Interested in a real-life example? Check out my marketing podcast, Marketing Over Coffee!



If you enjoyed this, please share it with your network!


Want to read more like this from ? Get updates here:


Marketing Blue Belt Preorder

Subscribe to my free newsletter!


How to track offsite conversions with Google Analytics

Michael Mindes of Tasty Minstrel Games asked:

Tasty_Minstrel_Games_on_Twitter____cspenn_I_heard_you_on_Social_Media_Examiner_podcast__Is_there_a_good_way_to_track_Amazon_Affiliate_results_with_Google_Analytics__.jpg

This is a terrific question and not necessarily as complicated as you might think. It requires some logic and inference.

First, begin with understanding what the goal is. On Amazon, the goal is sales. We want to sell as many of our products as possible. If you’re in Amazon Associates, then your goal is selling other peoples’ stuff. If you’re in KDP or Webstores, it’s selling your own products. For example, all of my books are sold both on my website and on Amazon.

Once you know what the overall sales goal is, you have to get your sales data from Amazon. Let’s use Marketing White Belt as an example. In the last 90 days, I’ve sold approximately $60 of Marketing White Belt on Amazon. So far, so good. I know what my traffic to Amazon is worth.

The next question is, how much traffic have I sent to Amazon? Using Google Analytics event tracking, I can track how many times people click on my Marketing White Belt book ads on my website:

Top_Events_-_Google_Analytics.jpg

89 clicks. Thus, I can make the starting inference that my value per click is $60/89, or 67 cents per click on my Marketing White Belt ads. Thus, I can set a goal value in my Google Analytics admin that any time I get an outbound click on my Marketing White Belt book, call it 67 cents of revenue.

Obviously, this isn’t exact. People can buy the book from Amazon without ever having been to my site. Thus, it’s important to rebalance. I’d run this analysis every 30 days and recompute the value of a click from my website to Amazon. Over time, I’d get enough data to create a reasonable average, and then use a rolling average to settle in on a value per click.

To Michael’s question, what about Amazon affiliates, where you’re sending clicks to other people’s stuff? Again, the same general logic applies. You know how much money you earn from Amazon each month, from the Associates reports. You should know, using Google Analytics, how many clicks you’re sending to Amazon. Work out what a value per click is, and you’ve got the beginnings of decent estimation.

Get as granular as you can, too. If you’re an Associate focusing on several different verticals, consider setting up event tracking categories. You might have one tracking event for electronics, another for books, etc. and then from your Associates reports, break out the fees you earned in each category. Now you can set up goal conversions per category.

Good luck tracking! Be sure to read up on Google Analytics Event Tracking.


If you enjoyed this, please share it with your network!


Want to read more like this from ? Get updates here:


Marketing Blue Belt Preorder

Subscribe to my free newsletter!


Predictive Analytics for Social Media at SMMW15

One of my complaints about marketing conferences is that the content tends to be the same old thing, over and over again. That’s understandable and even necessary, thanks to the churn rate of people coming into marketing. There’s an evergreen need for 101 content, for how-to, for the basics. Of all of the books I’ve written, the one for beginners, Marketing White Belt, consistently tops the sales charts.

But for those folks who have been around for a little while, conferences can get a bit stale. That’s why I’ll be trying something different at Michael Stelzner’s Social Media Marketing World this month. My session will be about social media analytics. Nothing new there on the surface, right? But instead of things you’ve already heard and done, we’re going to try something different together: advanced social media analytics.

Slide31-smmw15.jpg

What constitutes advanced analytics? First, we’ll examine a newer social media funnel that lets you characterize different metrics in a logical flow. That alone will help some marketers present more impactful reporting.

Second, we’re going to spend a lot of time on predictive analytics. There are three statistical patterns we’ll learn together: breakouts, trends, and anomalies:

  • Breakouts are changes in averages. When something breaks out, it experiences a significant and potentially lasting change. You had an average of 24 URL clicks per day on your tweets for the last 6 months. Suddenly, your daily average goes up to 36 clicks per day and stays there for a little while. That’s a breakout.
  • Anomalies are statistically significant aberrations. Your median number of engagements per day is 40. One day, you have 80. Is that significant? What about 180? We’ll look at how to tell the difference.
  • Trends are patterns in your data. Every day, you have one more person sharing your social updates than the previous day. Is that a trend? If so, where’s it going? We’ll study that.

With tools you already have or can afford (and by afford I mean as much as $30/month), we’ll see how these three kinds of analysis can help you predict the future. Once you know how to predict the future, you’ll know whether you want to keep it or change it.

Finally, we’ll walk through 3 recipes for predictive analytics together that you can take home and start using. The theory is great, but the take-home utility is even better.

If this sounds intriguing, then please do join me at Social Media Marketing World. Tickets are still available, both for the in-person event and for the sessions streamed virtually.

Disclosure: Registering through those links earns me a small but nonzero monetary gain through an affiliate program.


If you enjoyed this, please share it with your network!


Want to read more like this from ? Get updates here:


Marketing Blue Belt Preorder

Subscribe to my free newsletter!