A “study” making the new-media rounds illustrates the extent to which Internet radio boosters will go to cast commercial radio in a negative light. The study also inadvertently illustrates why pure Internet radio is failing and will continue to fail to attract any measurable audience.
A monitoring service, streamSerf, claims to have compared the number of artists played on terrestrial and Internet radio stations last month and found that Internet stations played 804,572 more artists than terrestrial stations (defined as commercial plus non-commercial stations).
There are several methodological issues that concern us about this study. We’ll get to those in a moment, but let’s run with the results for a moment. Here’s how audio4cast characterized the results:
It's not shocking to learn that Internet radio is more diverse. There are more choices for the listener, and more determination by the stations themselves to provide alternative music to the basic cookie-cutter formats and playlists provided on broadcast radio. It is stunning to learn that Internet radio's list of unique artists is greater than broadcast radio's by 3600%.
Basic cookie-cutter formats and playlists?
Let’s reflect on that phrase. Remember satellite radio? Beyond AM, beyond FM? We were told that commercial radio had become predictable and stale. Satellite radio with nearly 100 channels of music would reinvent radio, offering more diversity, finding new artists.
The one and only time that Arbitron measured satellite radio, it found that virtually no one listens to the majority of the channels. There are many channels where there are more people in the studio than people listening. The only channels that attract significant numbers of listeners are those that resemble commercial radio stations. There is a listener-driven reason for playlists. There is a listener-driven reason similarly formatted stations sound similar.
There is no reason to expect successful Internet radio to be any different. The vast majority of Internet radio listeners tune to stations that play the same music played by commercial radio stations. We can see this from the study’s own results. Look at the list of the top 10 terrestrial and top 10 Internet artists (click to enlarge).
Also interesting is the list of artists that get the most plays on broadcast stations versus Internet radio stations. While some of the top ten artists are the same on either list, others are very different.
The truth is that there is only one artist on the Internet list that stands out. Amin Van Buuren is ranked 6th on the Internet, but 7,811 on terrestrial. The other top ranked Internet artists include Madonna, Metalica, Lady Gaga, The Rolling Stones, and similar mainstream artists.
So having shaken off the shackles of basic cookie-cutter formats and playlists, successful Internet radio stations end up...playing the same artists heard on commercial radio stations.
But these are just the top 10 artists. What about the other 829,961 artists Internet stations play? According to the company, terrestrial radio played only 25,399 artists. Since thousands of radio stations have been testing their music now for nearly three decades, it is a safe bet that we have identified virtually every song that listeners want to hear.
If you have tested your music for any length of time, looking for new and different songs to play, you know that there are a lot more stiffs than hits.
We have found the same thing in every format, from Classical to Hip-Hop. Across the entire spectrum, there are only so many songs worth playing (if your goal is to have a measurable audience). That means there are only so many “marketable” artists that listeners want to hear.
We at Harker Research have been testing music for thirty years now. We find it hard to believe that there are 25,399 artists worth playing on the radio. We can only assume that the inclusion of educational and non-commercial stations ran up the numbers.
The study claims that Internet radio plays 3,600% (actually 3,168%) more artists, and pundits see it as a big plus for Internet radio. What it actually means is that Internet radio stations are playing over 800,000 artists that people don’t want to hear.
Before the Internet, artists that didn’t deserve to be heard beyond their family and closest friends toiled away in garages and empty coffeehouses. Years of practice would separate the good from the bad. The artists that got better found success. The majority went back to their day jobs.
Now anybody at any level of accomplishment can be heard on some quirky Internet radio station someplace. We suspect that the actual number of artists played on Internet stations is considerably higher than the 800K+ that the study found.
Paul Mockenhaupt, founder of the company, declared: It's the new, fresh, undiscovered, local, home grown music that's filling the internet airwaves! One report’s author adding: Internet radio gives voice to the long tail of music, providing entry for many musicians that have never had a platform before. That, he (Mockenhaupt) says, is the "magic" of Internet radio.
The magic of Internet radio is that it operates in a fantasy world. The notion that radio can operate in a “long tail” world is a fantasy as long as radio (whether it be terrestrial, in outer space, or on the Internet) depends on listenership as a measure of success.
But what about those numbers? How can a monitoring service be so precise? The reports are rather vague on the methodology of the study, streamSerf has no mention of the study on their web site, and we got no response to the questions we put to the company.
The web site boasts that it monitors thousands of radio stations:
Our patent-pending technology allows us to listen to thousands and thousands of radio stations at the same time... Satellite, Internet and Terrestrial stations. As a matter of fact, we are currently listening to over 30 MILLION HOURS of radio per year in 65 countries across the planet. If there's a song being played on the radio, there's a good chance we hear it.
It certainly sounds impressive, but let’s do a little math based on some pretty basic assumptions. We suspect that the company is monitoring station streams, not their signals. There are 14,355 licensed radio stations in the US, many of them not streaming. There’s a little less than another hundred on Sirius XM. Those are the easy numbers. How many Internet radio stations are there? Of those, how many are exclusively Internet? There are thousands, if not tens of thousands.
Thirty million hours of monitoring sounds like a lot, but that has to be spread across many radio stations. There are 8,760 hours in a year, so that means the company on average is monitoring 3,425 stations each hour of the year.
Considering that there are nearly 10,000 US FM stations alone, the company can’t be monitoring terrestrial radio stations continuously. It may be monitoring stations for a few hours at a time, or monitoring a handful of stations continuously. In either case, it means the company is taking a sampling. How the sample is chosen can have a major impact on the outcome.
More problematic is how one monitors tens of thousands of Internet stations along with the terrestrial and satellite stations. As with terrestrial stations, the only practical way is to either monitor a random cross-section of Internet stations continuously, or monitor every station known for just a few minutes.
Based on what we see, it appears that the 25,399 versus 829,971 artist comparison is more likely based on an educated guess and wishful thinking than one based on a month long continuous monitoring of all terrestrial and Internet stations.
It is making the rounds because despite its flimsiness, it provides one more opportunity for new media pundits to dump on commercial radio. The good news is that if the numbers are based on crude estimates, maybe terrestrial radio isn’t playing 25,399 artists. We hope not.