By RODNEY HO/rho@ajc.com, originally filed August 1, 2011
Pandora, the most popular on-line only "radio"-style option, is now as popular as some of the biggest stations in Atlanta among younger listeners, according to data compiled by Edison Research.
This is the first time such information has ever been released for Pandora publicly.
The ratings were very similar across the top 10 markets Edison studied.
Atlanta garnered a 0.7 average quarter hour rating for 18 to 34 year olds and a 0.5 AQH rating for 18 to 49 year olds. That's the same as New York, Chicago and Philadelphia. San Franciso, D.C. and Los Angeles had the highest ratings of 0.9 and 0.6 respectively in those two demos.
This would place Pandora in the same league as Hot 107.9 and Q100 for that 18 to 34 demographic.
AQH is different from the share numbers I normally get. It represents the percentage of people listening to that station during any 15-minute period. If you listen 5 minutes out of those 15, you get credit. A 0.7 AQH is equivalent in Atlanta to about a 9 share in that 18 to 34 demo. According to Arbitron, Q100 in June was third with an 8.6 share among 18 to 34 year olds while Hot came in second with 8.8. No. 1 is V-103 at a dominant 11.7.
Kurt Hansen, who runs RAIN, an online newsletter which has studied online radio for more than a decade, said this data will help Pandora get more local ad buys. Currently, Pandora appears to be mostly attracting national advertisers, he said. It needs advertising revenue to cover the costs of paying for royalties as more listeners listen to more music on the service.
Pandora listeners, of course, are all over the map, creating stations of their own choosing. For the unitiated, the concept is simple: pick a song or artist. Pandora, which has human beings define each song by multiple "genome" criteria such as musical style and beats per minute, will find songs of similar ilk. You can like or dislike those picks and over time, the station will become more attuned to your tastes.
The company, which struggled over the years to find funding as it gained fans, found momentum when smartphones became popular. It went public earlier this year.
[UPDATE 8/1/11. I was sent this interesting memo from Katz Media Group, which provides services to more than 3,000 radio stations and has a stake in supporting them, gave this counterpoint to Pandora's release:
The latest release posted on Pandora's investors' website is another attempt to position Pandora as something it isn't – a radio station. I understand their radio envy – they even refer to themselves as WPAN -- since Pandora is essentially a playlist maker set to shuffle. But what we took away from their announcement was itsfailure to reach half the country, fuzzy math, and no benefit for advertisers. So when it comes to Pandora comparing itself to broadcast radio, it just doesn't WPAN out.
Pandora's release centered around its having a strong presence in the top ten markets. But if the Pandora numbers are valid, then Pandora is essentially announcing that about half or more of its total listenership is restricted to the top ten markets in the U.S. Considering the top ten markets account for only about 25% of the U.S. 18-49 population, that leads us to the conclusion that Pandora is not being embraced outside these ten markets.
On the numbers front, Pandora is making claims based on what I'll call "Pandora's 'Special' Math." The companyhired well-respected Edison Research to create "average quarter hour" (AQH) ratings for "Pandora Corporate" (as they are listed by Triton Digital) within two age segments in 10 markets in the U.S. We aren't challenging Edison's math – they're only crunching the data Pandora gave them. But we do wonder about all the data that seems to be missing. Like the geographic parameters of their ratings. In order to use the only accredited ratings available, the numbers must be reported in terms of TV geographic definitions (or DMA). If so, that would inflate their ratings when compared to the smaller Arbitron radio metro area.Therefore, they cannot be making an apples-to-apples comparison. In addition, we're missing a statistic that's critical for any real radio station -- the weekly cume of their listening. As to why that was left out, we've pointed out before that the average Pandora listener tunes in just 2.5 times a week – clearly there's not much of a cume to be generated by that.
Another example of Pandora's 'Special' Math: In its initial analysts' call, Pandora claimed that it "ended 2010 with 2.3 percent market share of all radio listening in the United States. Six months later, Pandora has increased its market share to 3.6 percent." But anyone who looked at the numbers would see this was a mathematical impossibility. Triton Digital's published ratings show a net increase of 6% in users for Pandora over that six month period. For Pandora to equal 3.6% of all broadcast listening, broadcast radio listening would have had to drop by a third -- 32.5% -- in that same period. But it didn't -- In fact, broadcast radio listening remained consistent over that timeframe and, in the top ten markets Pandora targets, radio's AQH has increasedby 3.1%. And calculations by the two actual ratings sources for radio put all digital listening at about 3% of total radio listening. Without Pandora's 'Special' Math, Pandora would account for possibly 1-1/2% of all radio listening.
As we've said before, Pandora is not the ideal vehicle for advertisers. Let's look at ad targeting – which is crucial when a buyer is considering which radio stations to include on a buy against 18-49 year olds. Broadcast radio offers targeted environments where advertisers' messages will have relevance and meaning. A rating without an environment is meaningless. Pandora doesn't offer an environment, period, let alone a targeted one. So, for example, if the advertisement is for a Closed Circuit Boxing match, the spot could easily be surrounded by ballads from Celine Dion. Not what that advertiser signed up for, I bet.
Knowing radio's place in the hearts, minds and daily lives of hundreds of millions of listeners every month, we don't blame Pandora for wanting to be a radio station. But as the data continues to bear out, it's far from succeeding in making that case.
Mary Beth Garber
EVP/Radio Analysis and Insights
Katz Radio Group
7/28/2011
]
Join my Facebook fan page and Twitter.
By Rodney Ho, rho@ajc.com, AJCRadioTV blog
