Proof that hearing Hotel California repeatedly will drive you crazy

[michael hagerty]

Is this statement about focused on ratings or focused on revenues?


Would you say that means that #10 cannot make more than #1, ever? Because having the most listeners does not always mean making the most money.

Focused muscially is what I meant, but that translates to the other two.

Remember that you're not so much gunning for "the most listeners" as the highest share of listeners in a demographic the advertiser cares about.

#10 is a tough sell because most agency buys only go a few stations deep. Sometimes being tenth in the demo, you'll squeak onto the buy. Being #1, you're on the buy, and you'll likely be increasing your rates to compensate for having more demand than ad inventory. So yes, you'll make more money. Now, that's gross advertising revenue. If we're talking about making money in terms of net profit, then all sorts of factors come into play (how many people you employ, how much you pay them, any debt service from buying the station).

 

[davideduardo]

And if this is true: then, Pandora's model works better. Because at any given moment I can up or down vote any song that comes on.

Radio can look at MScores on individual song plays, too. In other words, how many people tuned out when a particular song played. They can track the tuneouts, also. It's another tool radio has.

As I just said above, Pandora may be a personalized music service, but radio in effect TRIES to be personal by selecting the songs it thinks the majority wants to listen to.

The Pandora offerings range from preprogrammed stations to totally personal channels, and the total array of possibilities is literally infinite. It's also national and catering to a slightly different listener base than OTA radio. But the big difference is that Pandora offers millions and millions of possible "channels" while each OTA station offers only one. One-to-one vs. one-to-many.

( a local focus group? really? )

Major market stations (basically the top 100 markets or so) do their own local research if they can afford it. That means stations in Huntsville and Panama City and Tucson and Bakersfield as well as ones in LA and NY and Chicago. If they can't afford it, they share projects between comparable markets. For example, some years ago when I was responsible for WTNT in Tallahassee, we did two tests a year shared between Albany, GA, Tallahasse and Dothan, AL. Each market had a test every 18 months, and we shared the results for the other tests. So my budget was for 2/3 of a test a year, but I got two full tests due to sharing... and the markets had almost identical tastes in country so it worked.

And the tests were local or within a hundred or so miles of us each time.

but mandated from a corporate office that has an agreement with the record labels to push certain songs/artists despite what that "local focus group's" findings reveal).

And your evidence of this is?

There are songs I didn't like the first 10 times I heard them, but after much repeating I have found myself tapping my toes to it. "Tell a lie often enough, it becomes truth comes" to mind.

Decades ago Guy Zapoleon said in an interview that you could not get reliable callout results on a song before it had about 100 to 125 spins. Other CHR programmers like Berkowitz confirmed this concept, saying that a listener had to hear a song at least 5 times to be able to give a reliable score.

And that is why cuurrent based stations do some form of current music research on an ongoing, continuous basis.

Aha, so the rules can change but the stakeholders have to agree to the change. In other words, if currently 10 seconds of a song is played in focus groups and that is what the stakeholders agree to.. to change to 15 or 20 seconds, all stakeholders must agree. You can't have a rogue PD thinking, "hmm I wonder if 20 seconds of a song will change the outcome of the results" without having it approved by the Industry as a whole.

When you use electronic dials or do tests online, you find that essentially everyone scores a song by the end of 5 seconds. That's why test hooks are pretty standardized at 8" each. If you do longer hooks, the participants get fatigued, bored and stop actively scoring and you can observe that they are frustrated. Online, testing systems move on to the next song as soon as the previous one is scored. Average scoring time is under 5 seconds.

And once a person marks or dials the score, they don't change it.

Because, as is, OTA radio will eventually lose ground.

It already is losing, but that is why nearly every significant radio group has a strategy to move to new media platforms over time.

There is much more to radio than any given station having the most listeners, because obviously there can be only one. There is much more to radio than playing songs that "test well", because there are still people who like songs that don't test well. They just may not have been picked for your focus groups.

Again, focus groups are not used for music research.

And if there is not even 1 person out of 100 that likes certain songs, then the group that likes the song is very small and irrelevant. But in reality, there are plenty of songs in a music test that score well with a small group, but are disliked by the majority. They don't get played. What gets played are the songs that all the subsets agree on.

There is much more to radio than knowing the focus group likes one song over another because one group may say "good song" and another may say "bad song". There is much more to radio including agreements between owners and promotional companies, including record labels. There are relationships involved and history too.

As I said, I've done well over 1000 music tests... about $25 million dollars worth... and I have never seen the implementation of a test be overridden by the name of the record label or anything similar.

 

[oldies76]

My point is that unless you actually ask EVERYONE what they like and don't like, you cannot know 100% for sure what the majority likes. It's kind of like the electoral college.

Exactly what I've been trying to convey, but the pros disagree. Radio uses "statistical" sampling to determine what "everyone" should like, and I totally disagree with that method. David brings up replication results. An example here: If 3 million people listen to KRTH and it's 400 song playlist during any given week, how can a quarterly music test, which only involves a hundred or so participants, determine what the other 2,900,800 people should listen to? It's nonsense. I bet those other 99.999% of the LA population have their own "other" classic hits favorites that NEVER get aired. What about those songs and personal favorites??

So in turn, these couple hundred music test participants are dictating what the other millions of listeners should listen to, based on their results. They are saying, play "Hotel California" 5 times a day or "Brown Eyed Girl" 3 times a day, when in fact the other million might say....Hey, why doesn't KRTH play the songs I like??

This has been a constant hot debate since KRTH reduced their playlists by the hundreds since the early 1990's.

And I would like the RD Manager to reflect on this, since I already know what the pros will respond.

 

[BOZ Profit]

As I said, I've done well over 1000 music tests... about $25 million dollars worth...

David, I know you have spent months (cumulatively speaking) explaining and reexplaining... And you have provided some of the most valuable information on the radio business on this site. So everything you have said and explained to each member your respond to is much appreciated.

I also appreciate that you took time to break down my comments which I think are some of the key points of contention for many of those debating you. Whether it is acknowledged or not, we will have to wait and see... Playing devil's advocate is not easy but it's fun.

For some reason, it is difficult for some people to accept that radio is just like any other business with market research etc. Does anyone argue that candy placed at kids eye level sells faster than candy up at the adult eye level? Or placing inventory you want to move fast on an end cap will move faster than in an isle? So it is with radio.

Maybe because radio has been that close friend, and some people are learning some things about the close friend they don't like or just didn't think could ever be. And they don't want to accept that information they learned.

 

[oldies76]

Stale? No. On target? Yes. KRTH is at the very top of the rating, #1 or #2, for the last two books. People like it. Two and a half million people like it. You don't. So listen to your music collection and stop making the same complaints, over and over and over and over and over and over and over and....

Yes it has been stale, DE.....The staleness began when Mr. Hamilton left office. YOU just do not realize it, the public does.

 

[TheBigA]

YOU just do not realize it, the public does.

Which "public?" The public has voted to make it #1 in LA.

 

[TheBigA]

If 3 million people listen to KRTH and it's 400 song playlist during any given week, how can a quarterly music test, which only involves a hundred or so participants, determine what the other 2,900,800 people should listen to?

But KRTH has had experience with larger playlists, and it hasn't brought the results they have now. RESULTS MATTER. You keep overlooking the fact that we have statistics on radio stations with large and small playlists, over a 30 year period, and we've seen the results over and over.

 

[BOZ Profit]

My point is that unless you actually ask EVERYONE what they like and don't like, you cannot know 100% for sure what the majority likes. It's kind of like the electoral college.

And I would like the RD Manager to reflect on this, since I already know what the pros will respond.

I think the pros on the site acknowledge this.

It comes back to the point that David has made about having been involved in 1000's of music tests. He has seen results first hand and they have not really varied much. Human behavior is actually quite predictable when you are in the behavior monitoring business. Sure, there are exceptions as with everything. But the research is not to find the exceptions to cater to. It's to find the "norm". What will most respond to?

Here is an example: if you put 1000 people in room and gave them 3 kinds of ice cream to taste: vanilla, chocolate and strawberry. The majority will say they like vanilla the most. But what about those in the room that would say they like cherry the most, if it were a choice? Is it not fair that cherry was not represented? Do the majority in the room care that cherry was not an option too?

The majority of people may even be thinking "why doesn't KRTH play X" (where X is any song, and that can be thousands of answers...) but ultimately they are happy because the songs that are played, they are happy with.

KRTH, or any station, is not really concerned that we will sometimes turn off the radio to listen to an iPod. That (turning off the radio to listen to something else) has been happening since vinyl was created ... they know that in order for people to turn it back on there has to be something we want as a majority.

I'm not sure I can simply it any further. Do I want CHOM here in Montreal to only play music I want? Sure. But it is not going to happen. There is a much larger audience to consider, and I have to accept that 1/4 of the time I will be thrilled that they play music I thoroughly enjoy, 1/4 of the time I will be happy because I do like it or I have been exposed to something new and like it, 1/4 not really care it's just become background, and 1/4 turning the dial or turning the radio off because I really didn't like what was played. And so it is for each listener. So radio tries to mitigate the 1/4 where we will go away, but it is inevitable, especially with all the options. But I do come back. Why? Because I know they will eventually play something I love, and they also provide other content I am not going to get anywhere else.

 

[BOZ Profit]

Focused muscially is what I meant, but that translates to the other two.

Remember that you're not so much gunning for "the most listeners" as the highest share of listeners in a demographic the advertiser cares about.

#10 is a tough sell because most agency buys only go a few stations deep. Sometimes being tenth in the demo, you'll squeak onto the buy. Being #1, you're on the buy, and you'll likely be increasing your rates to compensate for having more demand than ad inventory. So yes, you'll make more money. Now, that's gross advertising revenue. If we're talking about making money in terms of net profit, then all sorts of factors come into play (how many people you employ, how much you pay them, any debt service from buying the station).

Now the question becomes: What comes first the chicken or the egg? The format or the advertiser? So, how does a station choose a format or does it choose which advertiser it wants to cater to first? After all, some demos have more money than others. Some demos are more likely to spend their money than other demos. Some advertisers cater to everyone (toothpaste, toilet paper) but it is usually a specific person buying it for the household (the women). Why would a station cater to the demo of any small industry when the bread and butter is to cater to industries that serve the masses? Who spends more money? Men or Women? Why would any radio owner do a format that did not cater to the largest advertising industry, no matter the format?

 

[davideduardo]

Exactly what I've been trying to convey, but the pros disagree. Radio uses "statistical" sampling to determine what "everyone" should like, and I totally disagree with that method.

But you have never programmed a radio station, so you should take into account the opinions of those on these boards who have experiences. I can fairly safely say that every one of us with some programming experience has learned more from both our own mistakes and by observing the failings of other stations to have developed a broad overview-like "sense" of what works and does not work.

David brings up replication results. An example here: If 3 million people listen to KRTH and it's 400 song playlist during any given week, how can a quarterly music test, which only involves a hundred or so participants, determine what the other 2,900,800 people should listen to? It's nonsense. I bet those other 99.999% of the LA population have their own "other" classic hits favorites that NEVER get aired. What about those songs and personal favorites??

If you test multiple samples concurrently and get the same results, you can test more and more people and you will also get the same results but at extremely higher costs. The purpose of doing replication studies is to see what the bare minimum sample size is whereby you get results that will always replicate and which represent the entire universe you will project the results into. But a replication study also shows that sampling more and more people does not add to the accuracy of a test... or any research.

The objective in sampling is to find the smallest sample size so as to not waste money while accurately determining the behaviour of the universe. The smaller the sample, the less the research costs and the more often you can do it.

The simple fact is that if you test 100 people and only a 2% like a "Honey", you are not going to play it. You can sample 100 or 100 more and you will find that, still, only 2% of percent of that larger sample like that song. You still are not going to play it. And if you add more sample and a song that nobody in the first 100 liked, while 2 people in the second 200 like, you are not going to play it, either. You are looking for songs that nearly everybody likes or is at minimum neutral about.

So in turn, these couple hundred music test participants are dictating what the other millions of listeners should listen to, based on their results. They are saying, play "Hotel California" 5 times a day or "Brown Eyed Girl" 3 times a day, when in fact the other million might say....Hey, why doesn't KRTH play the songs I like??

The number of daily spins is not a direct result of the music test. That is a program director's decision. One PD might say that the top 50 songs on the test should play 5 x a day, while another might say that the top 100 songs should play 3 x a day. The interpretation of how to use the data to determine rotations and, even, the cut-off point, is in the hands of the programmers, not the music test or the company that does the test.

In the case of KRTH, Rick Thomas used his experience and the data about PPM time spent listening to determine how often each listener on average should hear a given song.

This has been a constant hot debate since KRTH reduced their playlists by the hundreds since the early 1990's.

Implementation of research is as individual as each PD. Were it otherwise, we would not really need Program Directors. And the ability to do this stuff is why some PDs make high 6-figure salaries and some don't.

What you don't get and apparently don't want to get is that music tests are just a tool for the programmer to use to build a station and a format. There are many, many variables. Music testing eliminates the variable of whether a song is playable, today, or not. How the station is put together is a much bigger issue.

 

[davideduardo]

Now the question becomes: What comes first the chicken or the egg? The format or the advertiser? So, how does a station choose a format or does it choose which advertiser it wants to cater to first? After all, some demos have more money than others. Some demos are more likely to spend their money than other demos. Some advertisers cater to everyone (toothpaste, toilet paper) but it is usually a specific person buying it for the household (the women). Why would a station cater to the demo of any small industry when the bread and butter is to cater to industries that serve the masses? Who spends more money? Men or Women? Why would any radio owner do a format that did not cater to the largest advertising industry, no matter the format?

Decisions on formats are made considering a series of factors...

Is there a format hole?
Is there someone in a format doing a bad job of it?
Do listeners perceive that the station is doing it badly?
How big is the partisanship group for the format?
If the format is offered and well done, will people who like the music actually change station?
Is the format appealing in any part of the overall sales demo spectrum of 18 to 54?

The most common format research method is called an ATU, which means Awareness - Trial - Usage. It involves playing a montage of hooks of a number of blends of music, each representing a possible format. Respondents are asked how much they like each one, on a scale of "I'd listen a lot" to "I'd never listen". Then they are asked if any local station plays that kind of music blend, and if they say there is one, how well does it do it and how much do they listen to it. They are then asked if a new station came on with that blend, would they try it. And then, if they liked it, which stations that they currently listen to would they listen to less.

This is very subjective, and requires lots of interpretation as any research that tries to predict the future is "if-y". The real objective is to find out if there are blends that are liked, and whether existing stations serve that niche well or not. Putting together a viable format is a totally different matter as success depends not just on the right music but also on marketing, signal, airstaff, promotions, rotations, imaging and more.

 

[michael hagerty]

Now the question becomes: What comes first the chicken or the egg? The format or the advertiser? So, how does a station choose a format or does it choose which advertiser it wants to cater to first? After all, some demos have more money than others. Some demos are more likely to spend their money than other demos. Some advertisers cater to everyone (toothpaste, toilet paper) but it is usually a specific person buying it for the household (the women). Why would a station cater to the demo of any small industry when the bread and butter is to cater to industries that serve the masses? Who spends more money? Men or Women? Why would any radio owner do a format that did not cater to the largest advertising industry, no matter the format?

Good questions.

First, it's important to note that the amount of dollars spent in any given major market in any given year is not a subject of much mystery. Budgets are drawn up well in advance. Broadcast trades and the larger financial news media report on what the likely amount of ad spending in a coming year will be. That can change based on conditions, but only by a few percent.

In my experience, stations decide on a format by looking at the amount of money being spent by advertisers in the market to attract a given demographic, then examine the stations already dividing that ad revenue. Analyses are done as to the strengths and weaknesses of the players already there (including their own...do we have the signal to cover the market adequately?). Is there an opportunity to rank highly enough to be an automatic buy in that demographic? If so, are we displacing a competitor or simply reducing the size of the slices of the pie? If the latter, is the result a number profitable for us given the way we want to run the radio station? If not, are we willing to make changes to the operating plan (voice-tracking instead of live DJs, a lower promotion budget) to make it profitable? Could additional promotion actually displace one of the competitors? What would that cost? Would the added cost reduce or perhaps eliminate the profit margin?

If the answers aren't satisfying, you move on to another demographic and perform the same analyses. Sometimes you'll find that (and this is just for illustration), while less money overall is being spent in your market on 18-34 than on 25-54, your chances are better of making more money because there are fewer competitors, or it would be easier to displace one or more of them from the format.

Again, if the answers aren't satisfying, you move on. Some stations find themselves in what might be considered "niche" broadcasting (religion, ethnic), but the numbers are better than what they could get trying to elbow into the pond where the big dogs are drinking.

If every station went after the largest ad spending category, there would not only be vast underserved yet profitable demographics, but broadcasters would continue to research, to promote, to make their stations respond to the behaviors and expectations of the audience. In time, some of them would rise above the others in the ratings, the agencies would buy the top five, seven, maybe ten of them....and the rest would starve, forcing them to go find where they could make money, which would lead to something that looks remarkably like today.

 

[BOZ Profit]

Decisions on formats are made considering a series of factors...

Is there a format hole?
Is there someone in a format doing a bad job of it?
Do listeners perceive that the station is doing it badly?
How big is the partisanship group for the format?
If the format is offered and well done, will people who like the music actually change station?
Is the format appealing in any part of the overall sales demo spectrum of 18 to 54?

Playing devil's advocate again here...

I thought the sole motive was to earn maximum profits and from what has been explained here is that in order to have maximum profits you have to cater to the largest audience. If there is a format hole for Jazz but only a fraction of the audience would listen, why would any station owner even consider it? And it seems many stations have moved away from Jazz, with a few exceptions. Shouldn't every station be in the same format competing for the same audience (the largest demo) in order to maximize profits?

I have asked before, why even have formats? Just play what the largest segment wants to hear at any given moment. If the majority feel like hearing Achy Breaky Heart right now, play it! If the majority want to hear Private Eyes right now, play it! Now everyone wants the new one from Lady Gaga, play it! Why waste time with formats? Just play what people (the majority) want when they want.

 

[davideduardo]

I
I'm not sure I can simply it any further. Do I want CHOM here in Montreal to only play music I want? Sure. But it is not going to happen.

Now that you mention Montréal, here is a real anecdote.

In the late 50's, CJMS was one of my favorite radio stations in the entire world. I had a friend tape them on cassettes and mail them to me. I went to Québec several times to listen and observe the jocks and the music flow and rotations. I even visited a few times.

But I was not going to put their playlist on my station in Quito. Yes, I might pick up some good songs that I thought might work in my market, and discovered things from France Gall to Marie Laforêt to Joe Dassin that I could successfully program... but in different rotations. Simply, I took the elements of a very good radio station and adapted them to my needs. That's the role of a program director... often taking ideas and concepts from other places and other stations and adapting them successfully to a different station's needs.

 

[davideduardo]

Playing devil's advocate again here...

I thought the sole motive was to earn maximum profits and from what has been explained here is that in order to have maximum profits you have to cater to the largest audience. If there is a format hole for Jazz but only a fraction of the audience would listen, why would any station owner even consider it? And it seems many stations have moved away from Jazz, with a few exceptions. Shouldn't every station be in the same format competing for the same audience (the largest demo) in order to maximize profits?

That is the "how big is the partisanship group" question. A missing format with few partisans will not make the cut. Or, in the case of jazz, a missing format with few partisans all of whom are over 55 will never make it.

And selecting a format is more than just getting a large audience. Sports stations often get very low 12+ numbers, but deliver adult men very efficiently with little spillage. So... as was the case for many years with WFAN, you can be 15th in 12+ but #1 in market revenue. All those things are considered as a format change is a management decision, generally not the PD's call.

I have asked before, why even have formats? Just play what the largest segment wants to hear at any given moment. I

There are no "majority" songs. So each station picks a lifestyle / music style group and plays the songs that group agrees on. If you did anything else, you would have Banda El Limon and Florida Georgia Line on KIIS.

 

[BOZ Profit]

In the late 50's, CJMS was one of my favorite radio stations in the entire world.

First must note the current day CJMS is not the one you are referring to.

Next, how did you discover it?

 

[TheBigA]

If there is a format hole for Jazz but only a fraction of the audience would listen, why would any station owner even consider it?

I may be the only person here who has non-commercial experience, so I'll address that. You consider it because you're in a market where there's demonstrated passion for that music, and it will deliver passionate listeners who will donate money to hear that music. And that's why we have non-commercial radio. But it puts a level of responsibility on the listeners to put their money where their passion is. Because you're right...very few advertisers want to reach that audience, although it's up to the station to find companies who want format association, which is different from advertising.

I have asked before, why even have formats? Just play what the largest segment wants to hear at any given moment.

I think I addressed that the first time you asked it with the Garth Brooks example. But I'll add this: The fact that a company can own several stations in the same market gives them the chance to reach several groups at the same time. So as I said, no station will reach 51%. But if you have three stations that each get 5%, you will get more sub-groups, giving you a larger portion of the population.

At the same time, music has become a very polarizing thing. As one of the posters has said every song is somebody's favorite. If you have three stations, you can play different groups of songs and reach different groups of people. And you can play songs that you could not play for the other group. So the people are the ones who have created the formats, not radio. There was a time when a station like WABC played everything from Al Hirt to Johnny Cash to The Doors to Frank Sinatra. They called the format "mass appeal," not Top 40. Rick Sklar was doing what you suggested, and he addressed that in his book, "Rockin' America." But there were limitations to his format, and lots of songs that didn't get played. Once FM came along, stations had the ability to narrow-cast and better define what people liked.

 

[BOZ Profit]

I may be the only person here who has non-commercial experience, so I'll address that. You consider it because you're in a market where there's demonstrated passion for that music, and it will deliver passionate listeners who will donate money to hear that music. And that's why we have non-commercial radio. But it puts a level of responsibility on the listeners to put their money where their passion is. Because you're right...very few advertisers want to reach that audience, although it's up to the station to find companies who want format association, which is different from advertising.

The argument between the pros and enthusiasts here has been that the pros give the reason Hotel California is played so often is because it is a mass appeal song that the majority have indicated they enjoy. And it is played on several stations. So, logic tells me why aren't all stations only playing the only mass appeal songs? That is my devil's advocate position... here for the sake of "arguetainment" (Am I the first to use this portmanteau? Feel free to use, but PLEASE give proper credit! : ) ). [Edit: I Googled it, and see that alas, it has already been coined]

I think I addressed that the first time you asked it with the Garth Brooks example.

I must have missed it... can you point me to your first reply with the Garth Brooks example?

So the people are the ones who have created the formats, not radio.

I have to disagree here. I think marketers did; more specifically radio people with that marketing savvy.

 

[michael hagerty]

Something else for our moderator:

Ad buys aren't equal. When I say "the agencies buy 5 to 7 deep...if you're tenth, maybe you make it on some", that's not entirely the agencies' call. Each of their clients has their own budget. Some they can just throw a bucket of money at the top 7 stations 25-54. Others may require applying with an eye dropper to get through the year or the quarter...maybe they only buy the #1 or top 2 stations in the demo.

And demand dictates rates. If you've got more buys than you have ad time, you raise your rates. If you're struggling to get buys, you lower your rates in hopes of looking more attractive...especially to the clients who just decided they can't afford #1 this time around.

But...and this is an immovable truth: Ad dollars are not linear. KIIS-FM billed $61.6 million in 2013. Getting half of KIIS' numbers won't get you $30.8 million in ad revenue, or anywhere near it. There's a steep cliff and, as with record sales back in the day, there's less difference between #11 and last place than there is between #11 and #1.

 

[michael hagerty]

The argument between the pros and enthusiasts here has been that the pros give the reason Hotel California is played so often is because it is a mass appeal song that the majority have indicated they enjoy. And it is played on several stations. So, logic tells me why aren't all stations only playing the only mass appeal songs? That is my devil's advocate position... here for the sake of "arguetainment" (Am I the first to use this portmanteau? Feel free to use, but PLEASE give proper credit! : ) ).



I must have missed it... can you point me to your first reply with the Garth Brooks example?



I have to disagree here. I think marketers did; more specifically radio people with that marketing savvy.

If you'll scroll back a few, I answered why stations don't all play only the mass appeal songs.

And for clarity, it could be better said this way:

"Hotel California" is a song that the majority of likely listeners to KRTH in Los Angeles who are within the desired demographic share in common as a favorite. There are differences between listeners of stations, listeners to different stations and listeners in different cities.

That, however, is not why it is played so often.

It is played so often because the majority of likely listeners to KRTH in Los Angeles who are within the desired demographic have listening patterns and habits that limit their exposure to any given record. For the average KRTH listener to hear the song often enough for it to register as a song that KRTH plays (remember, that average listener splits his or her listening time among between 6 to 9 radio stations over a two-week period), KRTH needs to play it often.

More than 2 million people listen to KRTH in a week, but only about 40,000 are listening at any given moment. That requires repetition (just like advertising) to attain reach.