KVNS (1700 AM) flips to Fox Sports

[gar hi]

That's the whole reason for Top 40... play 30 to 40 songs, and no others, because people want to hear their favorites over and over again.

Hey, it would be great if they even played 30 or 40 songs on top 40 radio but the ones I've heard in various markets seem to play the same top 5 to 10 current hits and mix that with the same old constantly played songs from within the past couple years.

That song 'Somebody That I Used To Know' was great when I first heard it but after hearing it 80 gazillion times, I don't want to hear it again.



I've been searching the net to see if others feel as though I do about stations playing the same limited selection of songs and wow, a LOT of people wonder the exact same thing.



http://www.google.com/#hl=en&sclien...366&bih=673&bav=on.2,or.r_gc.r_pw.r_qf.&cad=b





Seems the general consensus is that what they play doesn't necessarily represent what the people as a whole really want.



This answer seems to really explain a lot...



http://www.sodahead.com/fun/why-do-they-play-the-same-songs-over-and-over-again/question-1522045/





Everyone I know who range in age from their 30s to their 60s and like different kinds of music from top 40 to oldies all say the same thing too, that there's no variety and the same songs are constantly played.

That's why so many are now finding alternative sources for their music because there is much more variety.

 

[crainbebo]

Oh, and don't forget the CROSSOVER stuff! Adele's "Rolling in the Deep" was on too MANY formats! CHR, Hot AC, AC, Alternative, AAA and Modern AC. I'm surprised she hasn't made a heavy metal version of that song ("WE COULD HAVE HAD IT ALLLLLLL!!!!! ROLLING IN THE DEEP!!! YEEEEEEAAAAA!!!!!!") or a country version with steel guitars, banjos, and Carrie Underwood duoing with her.

-crainbebo

 

[tomservo]

It appears the only way any of us are going to get our dream oldies station is if some rich old matron leaves us her riches and we start a station in the noncomm band with the interest on the money!

Also, if there was any truth to the "stations tell listeners what they want" we'd be seeing a mass migration off of terrestrial radio because there are now many alternatives: streaming out the wazoo, satellite radio, even DMX/Music Choice. Each has its proponents that swear it's the "radio killer" but collectively they haven't really put a huge dent in listening to broadcast radio at all. The truth is, broadcast radio IS still successful as a mass appeal medium; most of us just happen to be outside that mainstream.

 

[gar hi]

I had a job many years ago where they would play an adult contemporary station (which also played some older music in the rotation) in the background through the PA system and it drove me crazy having to listen to the same friggin songs every day.

Now, I work at home and at least I can change the station or listen to internet radio.

It makes no sense that anyone would want to hear the same songs over and over again.

There are many oldies and more recent songs I now wish I would never hear again but there was a time when I thought they were great songs.

Too much of anything is not good.

 

[davideduardo]

I had a job many years ago where they would play an adult contemporary station (which also played some older music in the rotation) in the background through the PA system and it drove me crazy having to listen to the same friggin songs every day.

Part of the issue there is that most AC stations are targeted at women. One I know quite well, in the very competitive LA market, did not even include men in its audience research.

I'm assuming you are not part of the target group of the station in question. So it is natural that you would be annoyed by some of the songs if not the entire station.

I'm not a fashionista. If I'm at the doctor's office and all the magazines in the waiting room are things like Elle and Vogue, I won't even pick them up... they hold no interest. But the publishers of those books will never adjust content to try to please me... it is just impossible.

Radio works the same way.

 

[rbrucecarter5]

In general, they prescribe what we're supposed to want and if that can be pushed on the public enough, they have no choice but to subscribe to it.

Here is a little illustrated article about how stations test music...

http://www.americanradiohistory.com/research_AMT.htm

Replication studies show the results to be accurate within a couple of percent, which is far more accurate than needed for the purpose.

Quote from the link:

"precisely recruited group of listeners or potential listeners"

Precisely recruited - I rest my case. They are picked to support a preconceived notion. The research, therefore, is fatally flawed from the start. The results will reflect exactly what the people conducting the survey want to happen.

Whatever the methodology used, it obviously doesn't work overall radio listenership is decreasing. People are dissatisfied, so they look for alternative. Even ones that cost, like streaming and satellite.

You cannot fit people into little neat boxes. They will dislike it. They will rebel and break out of those boxes. The people left will be wishy washy, willing to accept anything the stations put out because to them it is just background anyway - they are passive, non-committed, not loyal to the station or its advertisers. You are programming to lemmings, who are probably dial flippers / button pushers. The moment something else comes along, they will be gone. The lowest common denominator. Active listeners who strongly support stations and advertisers - GONE FOR GOOD.

Congratulations. You have programmed exactly what you want to listen to, and now have a jukebox playing your own greatest hits. So does everybody else, but with an iPod instead of a broadcast station.

 

[davideduardo]

Quote from the link:

"precisely recruited group of listeners or potential listeners"

Precisely recruited - I rest my case. They are picked to support a preconceived notion. The research, therefore, is fatally flawed from the start. The results will reflect exactly what the people conducting the survey want to happen.

"Precise recruiting" means getting a true proportional sample of the target audience of a station.

A music test asks people to score on a precise scale hundreds of songs by listening to snippets.

Every station has a core audience demographic. Using one I worked on for many years, it was adults 32 to 44. So if a person is recruited, they must be within that age group. Yes, there were listeners outside that narrow age range, but 80% were inside it so we used a bell curve principle to determine where the station had to do well... the rest fall into line based on satisfying the greatest sized group.

So the second key issue of recruiting is to get people who like your kind of format. You would not invite heavy metal diehards who despise country to a country station test, would you? So a recruit will either play "pods" of representative song styles from the general format (sub-genres) and if a person does not, let's say, like 3 out of 5 pods, they don't get in. Another way is to determine if a person uses your station or a direct competitor, and if they do and they do so for a minimum number of weekly hours, they get invited to the test.

So you invite people in your target age range, and people who like your station or your station's kind of music already. You then make sure you get enough people in each sub-age-group (maybe in my example 22-38 and 39-44) to be balanced.

What you have is a highly accurate mirror of your station's potential audience. Professional recruiters are used, and random sampling techniques are employed. The cost of recruiting alone is about $100 per person, and each person is typically given from $75 to $150 per test for showing up, plus a buffet meal or snack.

In radio, a couple of percent variance on each song's score is not significant... so exact precision is not needed. Still, a music test can run up to $50 thousand dollars each, and many larger stations do them several times a year.

Whatever the methodology used, it obviously doesn't work overall radio listenership is decreasing. People are dissatisfied, so they look for alternative. Even ones that cost, like streaming and satellite.

Radio's very slow decline in listening levels is due to a profusion of entertainment alternatives that never existed before. Still, around 94% of Americans listen to the radio, and in diary (smaller markets) the listening levels average around 17 to 18 hours a week and in PPM markets using a different methodology, they average around 12 hours a week for every person 6+ in the US.

Active listeners who strongly support stations and advertisers - GONE FOR GOOD.

Not true in any way, form or fashion, and provable by ratings data.

Congratulations. You have programmed exactly what you want to listen to, and now have a jukebox playing your own greatest hits. So does everybody else, but with an iPod instead of a broadcast station.

You have it backwards. The listeners are asked what they want to hear, and then that is given to them in an easy to use delivery method OTA, by streaming and by apps.

 

[MarioMania]

CHR is playing only Top 10 not Top 40 these days, Songs playing Over & Over & Over & Over & Over..really kills it for me....

 

[rbrucecarter5]

You have it backwards. The listeners are asked what they want to hear, and then that is given to them in an easy to use delivery method OTA, by streaming and by apps.

Why go to all that trouble? All it would take is a web site, a list of artists and titles that fit within the format, pick the top 1000 songs voted for by listeners - there is your rotation. Save a lot of money and time. I've already seen that on some radio station websites.

 

[davideduardo]

You have it backwards. The listeners are asked what they want to hear, and then that is given to them in an easy to use delivery method OTA, by streaming and by apps.

Why go to all that trouble? All it would take is a web site, a list of artists and titles that fit within the format, pick the top 1000 songs voted for by listeners - there is your rotation. Save a lot of money and time. I've already seen that on some radio station websites.

First, "list based voting" is not useful at all. You must have the listener sample actually hear a snippet of a song to get a real "intent to listen" score. We often think we like a song when we see the title and artist, but in reality we like some more and some less when we hear them than when we read them on a list.

Also, when confronted with a list, many people don't know the exact titles to many songs they like by artists they like. You gott'a hear 'em to score them. Horses win by running, not by standing in a stall.

There is no practical way of getting a proportional random sample of both people who listen and who might listen to a station on a station website as you suggest.

While website polls may make listeners feel empowered, and give stations a bit of additional data, this is not a usable data source.

First, the visitors to a station site tend to already be partisans. And they may feel prone to voting for the station and not the songs. Second, your competitors can poison the well by registering with spoofed usernames and vote high for the stiffs and low for the good songs.

But the biggest reason why this is not practical is that it is not anonymous. Listeners enter with a bias caused by the branded website. AMT's are not identified with a station, and are done by independent research companies using independent recruiters at a commercial meeting room.

Even research companies that do web-based music testing use an unbranded website so as not to influence the listener.

Finally, for any test that includes recent or current music, the test has to be done on all respondents over a very narrow window... at most, a week. Website polls tend to get fewer than the needed responses... and most people, voluntarily, do not sit through 500 to 1200 song clips unless incentivized.

Of course, there is also the possibility (actually, a high probability) that people will incorrectly identify age, gender and other characteristics on a self-recruited web test so that the data does not correctly represent the desired group of listeners.

 

[tomservo]

This topic's meandering into research has been interesting. I don't think I've ever shared online the fact that I was once invited to one of these research events for the local active rock station when I was in my 20s. I listened to snippets of songs over the phone and gave opinions (all negative, probably) and was still invited. At the event, they didn't do a song by song query, just asked general opinions. As I recall, they did not implement a single suggestion from that panel. They also showed us possible logo/branding changes, but didn't do any of that, either. Finally, they promised us pizza but only had enough for half the participants. :

I was never invited back.

 

[davideduardo]

This topic's meandering into research has been interesting. I don't think I've ever shared online the fact that I was once invited to one of these research events for the local active rock station when I was in my 20s. I listened to snippets of songs over the phone and gave opinions (all negative, probably) and was still invited.

The snippets were intended to determine if you were a possible or actual listener to a station. You obviously opened the recruit screener "gates" by way of your answers.

At the event, they didn't do a song by song query, just asked general opinions.

That would be a perceptual project, generally referred to as a focus group. Those projects are intended to get opinions on current programming, potential changes, new TV commercials, logos, talent, etc.

Typically, focus groups are 8 to 12 people, and a station may do a set of 3 or 4 at a time. A music test is 50 to 100 people in a session.

As I recall, they did not implement a single suggestion from that panel. They also showed us possible logo/branding changes, but didn't do any of that, either.

Without knowing what the other focus groups suggested or said, it's hard to say why you didn't notice any response to the group's observations.

Finally, they promised us pizza but only had enough for half the participants. :

Generally, such tests are done at a focus facility which is in charge of providing refreshments... and they screwed up.

I was never invited back.

Stations and research companies are generally pretty adamant about not having repeat respondents. That said, today in PPM markets where ratings are done with panels, stations are reconsidering using respondents over and over to be able to track preferences along a timeline.

 

[tomservo]

The snippets were intended to determine if you were a possible or actual listener to a station. You obviously opened the recruit screener "gates" by way of your answers.

That's interesting; I'm almost positive I disliked everything they played, because I didn't like the active rock format that much. Every song they played was in rotation at the station where I held panel with the other guests.

Generally, such tests are done at a focus facility which is in charge of providing refreshments... and they screwed up.

This was at the station's studio. Or, rather, the cluster's building. We got a tour of the station and saw where each of the group's stations were housed. It was interesting, in a bland corporate radio sort of way. The focus group was held in a corporate meeting room on the same floor as the station, hosted by the GM, PM and one of the better known jocks.

It was an interesting event for a radio geek like me, but everyone else seemed pretty bored with it. David, your insights into this sort of thing are also great. The business of radio is something else.

 

[rbrucecarter5]

It as an interesting event for a radio geek like me, but everyone else seemed pretty bored with it. David, your insights into this sort of thing are also great. The business of radio is something else.

Yeah - I agree - keep the insights coming, David!

It still seems, though, that the testing is not scientific. There is no double blind control group. It seems that each station is absolutely determined "I will stay in this active rock format no matter what. I will not play a single song outside of that format. I will find the 50 songs that fit my format, and these people will pick them and they will be perfect. I will not program to a single listener below 18 or above 34 - I don't want them. My demographic will be perfect and will match the mold. I will not be guilty on non-conformity. I will present to advertisers the perfect demographic and the perfect play list. Anybody not fitting into my stations mold is strange, outmoded, and does not deserve radio service." It all sounds a bit totalitarian, dystopic, and completely dominated by a fanatical paranoia about squeezing every last dollar from potential sponsors - the heck with the listener, creativity, and individuality.

 

[rbrucecarter5]

Very astute observations:

http://www.radioinfo.com/2012/09/26/challenges-for-smooth-operators/

The left out 95.7 in Houston, although that might have been just over the five year limit. Smooth Jazz is one of those HD-2 formats in Houston, as are oldies, Christian rock. HD-2 is one of those places radio stations send unwanted formats to die. They can say in good conscience "we are making the effort" - when in reality they know nobody listens to HD radio. NPR is having fits with enough listeners complaining they do the only thing they can - go after additional over the air outlets for formats with vocal advocates like classical and jazz. Too bad the commercial stations don't do the same with their unwanted formats.

 

[davideduardo]

This was at the station's studio. Or, rather, the cluster's building.

This is then an example of a station trying to do it's own research to save some money, i.e. a couple of pizzas instead of the normal $15 to $20 thousand for a set of three to four focus groups with a trained moderator at a focus facility (one of those with one-way glass) and a full writeup of the results with analysis.

And, it's no different than someone trying to build a nuclear powered home energy source... it's way beyond the ability of the average radio person to do recruiting, moderation and interpretation on their own.

Or, simply, "you get what you pay for."

 

[davideduardo]

It still seems, though, that the testing is not scientific.

Sure it is. The objective is to take a proportional and representative sample of the listener base to a station (or a format) and find out, within the format, what songs should be played and not played, and among the former, which should be played the same amount or more or less.

There is no double blind control group. It seems that each station is absolutely determined "I will stay in this active rock format no matter what. I will not play a single song outside of that format.

A music test is kind of like doing monitor points for a directional array; if anything is "off" you make small adjustments afterwards. But you don't change power, you don't change frequency, you don't change transmitter site after doing "points."

What you call a "double blind" is really what is called a "format search." In that process, a number of music genres are evaluated by a broad range of people in a market to see if there is an opening or opportunity. I've done format searches with as few as six alternative music styles tested, and with as many as 17.

But a station does not look for a new format if it is doing well... even just "moderately well". That's because a format change generally causes billings to reset to zero, and you are guaranteed operating losses for the year of the change due to very slowly building revenue and the higher costs of promoting a new format, personnel changes, etc. And you have no guarantee
that the format will work since the concept may be great, but the execution may not attract enough listeners.

I will find the 50 songs that fit my format, and these people will pick them and they will be perfect. I will not program to a single listener below 18 or above 34 - I don't want them. My demographic will be perfect and will match the mold. I will not be guilty on non-conformity.

No, that is not it at all. Let's use a mining analogy... you have a gold mine, and one vein is very productive and the extraction of gold yields a great profit. There are a number of smaller veins that you don't mine, though, because the cost of getting the gold out exceeds the value of the metal extracted and sold.

When radio stations do music research they go for the core listener in a format. Let's say a station is CHR... appealing to 12 to 34 year olds, predominantly females. The core for sales, though, is Women 18-34 and you can't get teen money in radio no matter how hard you try. So you research the 18-34 women, and to make sure the door is open to teens as they mature, and you include some 16 to 18 year olds. And to not be totally negative to men, you bring in males as a third of the 18-34 sample.

We know that CHR has under-16 appeal, and over-34 appeal but we also know that the key is to get the core... both to guarantee sales and to successfully program the station against competition. We know that if we get the 16-34 sample described we will get lots of overflow.

There is no need to sample outside the core because it could be dangerous: some songs that early teens like may be despised by the core and some songs the over-35's do like may be totally negative to the 16-24 segment... so the knowledge of what those songs are is not actionable.

I will present to advertisers the perfect demographic and the perfect play list. Anybody not fitting into my stations mold is strange, outmoded, and does not deserve radio service."

No, it's simpler: anyone who does not like the core songs and core artists of my format is not going to listen anyway, so I won't survey them.

It all sounds a bit totalitarian, dystopic, and completely dominated by a fanatical paranoia about squeezing every last dollar from potential sponsors - the heck with the listener, creativity, and individuality.

Radio stations use ratings to set prices. Agencies use ratings to hammer prices. But the ratings metric more or less determines how much a station can bill.

It's not about "squeezing dollars" at the program side; that is a sales function once you have ratings.

In programming, music research is almost always done to fine tune the existing format. You test the songs you play. You test some songs that you might play. You test some songs the competitor plays. You test a song or two that were in movies just in case. You try to test every song that your listeners would like. The end result is that you know, every time you test, what the core likes, and, using things like cluster / factor analysis, what the subsets of the audience like (here the subsets are not based on age, but on music tastes... an alternative station may have four or five dueling subests and you are looking to form a coalition based on commonalities).

Radio is a push medium, not a pull-based customized Pandora stream or a personal iPod playlist. The hard thing to do today is to produce a blend that is, in the long run, better and easier than the alternatives for many moments in listeners' days and lives. You do that by adjusting the existing product to keep it responsive to listener needs.

 

[rbrucecarter5]

What you call a "double blind" is really what is called a "format search."

No - I was talking about a scientific technique that eliminates bias in the results. I don't know exactly how you could apply it to radio research, but the results could be guaranteed completely objective if it was.

http://en.wikipedia.org/wiki/Double-blind#Double-blind_trials

 

[davideduardo]

No - I was talking about a scientific technique that eliminates bias in the results. I don't know exactly how you could apply it to radio research, but the results could be guaranteed completely objective if it was.

In the case of professionally conducted music tests, the elements that can cause bias are generally eliminated by the fact that the tests are recruited by a professional recruiter who does not identify the station, and conducted by a research organization that does not identify the station or company the test is being done for.

The classic double blind is done for things like drug testing where we know people often reply that the "feel better" if they take a pill even if they don't feel better. A music test is less subjective.... it is about whether you'd like to hear a song or not, and by how much.

Many researchers have done replication studies to define the ideal sample size. In such a study, a much larger than normal number is tested, and smaller groups are "assembled" by random number selection. These smaller groups are analyzed and compared; the objective is to determine with certainty the sample size where every time you test at that level the results are statistically the same.