Radio needs to Rethink Streaming
Clear Channel just announced second quarter results including this comment:
At iHeartRadio, we introduced an even more personalized listener experience with the release of iHeartRadio 5.0 and grew our registered users by 50% year over year.
Oddly enough, while numbers of registered users may be higher, consumption of Clear Channel streams has been treading water for the past year, and in May actually declined from the previous year as measured by Triton Digital.
But it isn’t just Clear Channel. There’s clear evidence that over all on-line listening to broadcast streams is peaking.
Take a look at the graph. It illustrates what’s going on.
We’ve totaled the Hours Tuned for Triton Digital’s Top 20 and plotted each month’s change from the same month of the previous year. The red line shows the combined trend for all services. The green line shows the trend of the Top 20 less Pandora.
Pandora makes up 78% of all listening in the Top 20, so the red line is a good proxy for Pandora.
Total streaming growth, as measured by Triton, peaked in July 2012 as listening doubled over 2011, up by 102%. Pandora listenership alone increased 151% over July 2011.
Since then the rate of growth of the top 20 services has steadily declined except for up-ticks in March, April, and May. Growth bottomed out at 9.6% in February, and even with the up-ticks, growth stands at 18% year-over-year. That’s including Pandora.
Without Pandora to prop it up, the Triton Digital trends look even worse. As the green line shows, in March of this year growth for the other 19 services stopped completely, and in the past two months listenership has actually declined.
That bears repeating. Growth for streaming services ranked two through twenty combined have less consumption today than they did one year ago. And it isn’t a fluke. The group has been in a steady decline since April of last year.
From May of last year consumption for Cumulus streaming is -26%, CBS -15%, Entercom -6%, and Clear Channel -1%.
Of course, Triton Digital measures many streams but only publicly releases the ratings of the Top 20 services. Is there perhaps a more positive story looking beyond these twenty?
Triton’s May report includes a graph of total listening hours comparing the last three years. Total listening is near 1.8 billion hours compared to slightly over 1.4 billion hours last year.
Most of the gains, however, can be attributed to Pandora’s spring spike. Earlier in the year 2014 gains were trailing previous years by a significant margin.
So the malaise we see in the Top 20 appears to extend to Triton’s weaker clients.
Taken as a whole, the numbers strongly suggest that streaming as measured by Triton Digital is reaching a peak. Even Pandora as dominant as it is appears to be maxing out.
How can it be that both broadcast over-the-air Nielsen numbers and Triton’s broadcast stream numbers are both eroding? Where are listeners going?
Triton only measures the streams of clients. There are many other streaming services that Triton does not measure.
The best guess is that unmeasured services like Spotify, iRadio, AOL, Rhapsody, and others are luring listeners away from both Nielsen measured broadcast stations and Triton Digital measured streaming stations.
If that’s the case, maybe just porting our stations onto the Internet isn’t the answer.
If people are looking for on-demand services, perhaps local broadcast stations need to develop distinctively different products for the Internet that combine local elements with on-demand elements.
Maybe the sweet-spot for local radio is a local Spotify-like station where listeners can create their own personalized local station?
The anemic performance of broadcast streams combined with Pandora and Slacker flame-outs suggest that we are far from determining what 21st century radio will look like.
The only thing we can safely conclude is that it won’t look anything like what we’re doing.