Current Issue:

Current Issue

On The Cover:
Cumulus Chairman Jeff Marcus

Click here to subscribe to Radio Ink.

Radio Ink Writers

Questioning The Pandora Numbers


From time to time we get criticized by readers for taking data and publishing it without drilling down or asking for more information. It's a fair criticism because all numbers can be massaged, stronger dayparts highlighted, or demos pushed out. After all, who's going to send out a press release that screams, "We came in 19th this month, please write about us."

One of the biggest struggles everyone is dealing with, when it comes to audio listening, is accurate measurement. How do you think the folks at ESPN feel? They are innovators when it comes to pushing listeners to their website and mobile or apps they've created. They will never get credit for those listeners when an ad buyer is looking at PPM data, yet everyone knows the numbers are huge.

Triton's monthly rankers have been consistently used to measure online listening. We do wonder though how many ad buyers and local advertisers understand how to incorporate "Average Active Sessions" and "Session Starts" into cash. Perhaps there are some, but its doubtful there are many. What radio -- or audio distribution -- needs is at least one reliable measurement service where over-the-air listeners can be combined with Web listeners and the combination used to increase rates. Those numbers for radio could be huge. Why shouldn't Clear Channel get credit for a Z100 listener in Texas who's consuming the audio on iHeartRadio? That listener is just as valuable to Home Depot or Progressive Insurance or Budweiser no matter where he or she lives.

Back to the criticism about publishing unchallenged data or research. Everyone noticed the Pandora numbers experienced a drop in June. As a result of that drop we received the following statement and questions about our coverage of the Pandora numbers from Vice President of Digital Digital Marketing & Research at Katz360 Maggie Hauck and EVP/Radio Analysis and Insights Katz Radio Group Mary Beth Garber.

Its interesting that Pandoras press release said its June ratings were great and showed growth. But the data posted in the Triton Top 20 syndicated ratings actually show a slight decrease in Pandoras usage in June. 

"So we come back to the question that Pandora has always declined to answer with any tangible data:  How do they calculate their share of 'total radio listening including satellite radio' if:
- There are no published reports that calculate the monthly total hours spent listening to satellite radio (let alone just commercial satellite radio).
- They cite Arbitron as a source but refuse to produce the number they used from its data (probably because they arent licensed to be selling with the information).
- Pandora tends to base its online share claims on just the Top 20 Stream Report by Triton. There are many more streamers that report to Triton but arent reflected in the Top 20 report. Which numbers do they use in their formula Top 20 or all Triton reporting streamers? And there is no way to calculate time spent listening to Spotify, lastFM, Rdio and other music streaming services that do not report to Triton, so they cant calculate all of online radio listening no one can.
- Could they please provide a link to the Census data table that supports their claims and indicate what data they are using from it in their formula?
- Why wont they simply show the components of the formula and cite the sources?

"Somehow, 'data from Arbitron, Triton and the Census' dont really answer any questions. Especially when those of us with access to all three sources cant replicate any of their claims. Are you really going to accept that vague answer from Pandora and be willing to continue publishing their claims without further documentation? Or will you ask them about these discrepancies before publishing their next set of claims?"

(10/25/2013 7:13:03 AM)
uEfURC This is one awesome article.Really looking forward to read more. Really Great.

- NY
(8/6/2012 3:56:53 PM)
Bud Longhorn = SMS (ba.broadcast) LOL!

- HDRadioFarce
(8/2/2012 9:24:53 PM)
Pandora is desperate to be called radio. It isn't.
Its a jukebox accessed by a wire and a computer.
Pandora is desperate to rake in the bucks, and can only do so by hocus pocus. In my opinion, its a con game. All we need are the facts, and they will not supply them. As far as Pandora's ability to survive,
genuine AM-FM radio will ultimately leave them in the dust where they belong.

- Bud Longhorn
(8/2/2012 8:17:57 PM)
Re: "If they want to show "real world" numbers, let them make public that report, it will show exactly how many hours of listening they have."

To Andy Collins:

And why would Pandora want to release these numbers? Do we see radio listening stats (or revenue figures) for CBS Radio? I hear no one calling for a release of number to prove Clear Channel's Q2, 2012 revenue report claims of "Attaining nearly 95 million downloads and upgrades of the iHeartRadio app, and totaling just under 112 million total listening hours in June, up 131% year over year..."

There are a number of groups that hold proprietary numbers close to the chest. It gives them an advantage of mystery.

Pandora owes nothing to the radio industry. It is using MRC accredited Triton Digital (along with other research firms) to quantify their stats to those who matter - the advertising community. I'm amazed that broadcast radio, while claiming how little Pandora affects its business, is concerned with Pandora.

A lawsuit would settle this claim once and for all... but how much egg on its face would the broadcasters have when Pandora's numbers prove credible?

- Ken Dardis
(8/2/2012 2:02:06 PM)
This reminds me of iBiquity's quoted number of HD radios sold. As one retailer was quoted, "the return rates on HD radios are higher than on other home products". Haven't heard any numbers recently. Lies have to be covered up with more lies, and one forgets ultimately what initial lies were told.

- Anon

Add a Comment | View All Comments


Send This Story To A Friend