Music - Special Programming

Doc:  I know you usually don’t give your opinion about things since you usually say something like, “Ask the listeners.”  However, I wonder if you could make an exception this time and give me your opinion on something.

I’m trying to come up with ideas for special one-hour or two-hour shows to put on during the holidays.  The shows don’t have to relate to any specific holiday—just something that is a little different.  By the way, the format doesn’t matter—I’m just looking for a few good ideas.  Any suggestions? - Anonymous

Anon:  You’re correct in saying that I usually don’t give my opinions about things because I know my opinions are worth a load of Vulpes Fulva leavings when it comes to your listeners, or listeners in any market.  When I’m asked what the listeners think, my motto always is: “Ask them.”

But I’ll bend the rule here a little and give my opinion about a special program.  But keep in mind that I’m not a PD or any other type of creative programming developer.  (You can’t criticize me too much since you asked for my opinion.)  How about this…

Most people like to sing along to their favorite songs, or even some songs that aren’t their favorites.  The lyrics to most songs are easy to understand and therefore are easy to sing along to.  But there are some songs that are virtually impossible to sing along to—either the lyrics are impossible to understand, or the lyrics are sung too quickly.

I think a unique show would be titled something like, “Songs You Can’t Sing Along With” or “Songs You Shouldn’t Pick at a Karaoke Bar.”    I’m sure the idea would work for any format, but here are a few songs that might be included in a show for some type of Oldies or Rock radio station:

Songs You Shouldn’t Pick at a Karaoke Bar

Along Comes Mary - Association

Thuderstruck - AC/DC

Boogie Woogie Bugle Boy - Bette Midler

And one of the all-time greatest “you can’t sing along with” songs is…

Goin’ Home - 10 Years After  (I wonder if Alvin Lee knew what he was singing, especially in the live recordings).

There ya go.  If the idea is interesting to you, I’m sure you can come up with several hours of songs.  You might even develop some type of contest where the winner would be the person who can successfully sing the lyrics to one or more songs you play.

By the way, you may also find some help for show ideas by going to music-related websites, such as Am I Right?, which is a great music site.

Music: Talking Up Songs

This may sound like a dumb question, but I'm going to ask it anyway. How long should a Top 40 station call music 'new'? Should this be based on the number of weeks/months/spins? Do reaction records require fewer talks, as they are more familiar faster? If your station is the only one playing a song in the market (but the song has been out for several months), is it fair to call it 'new'? It would be new to our listeners, but then again they may watch MTV. Is it OK to call a song the 'latest' from the artist? Thanks. - Anonymous

Anon: I have been involved with research investigating the word "new" as used by a radio station, but not for music. Therefore, I don't have any idea what average listeners think about this. I'm don't know the amount of importance listeners place on a song being called "new" or "latest."

While your question is important, I'm not sure if it justifies a separate research study. However, you should include it in your next study, or ask them at a remote, or maybe even ask listeners when they call the station (although that's not too reliable with all the potential problems with request lines).

Without the benefit of listener information, you might set your own rules and wait for listener reaction. In other words, you might use "new" for 30 days or so (the song may be on MTV, but it's new to your radio station), and then switch to "latest" after 30 days. If you think about this for a moment, this is really a semantics question. The terms are probably interchangeable for most people (even though you're using "new" and not "newest").

Music Taste Shifts

When we first started our Internet music research, our top 10 hooks reflected a very nice balance between the different styles of music we play. (We are CHR, and the top 10 averaged to be 30% Rhythmic, 40% Modern Rock-Pop, and 30% Pop). Lately the stats have our top hits consisting of 80 % Modern Rock-Pop, 20% Pop, and 0% anything else.

Our sample is roughly the same—same size, age, etc.  And our sample sizes now differ little from the sample sizes when we started. I should also add that the respondents volunteer to rate the songs. Any idea why a dramatic change in our results has occurred?

On a related, yet different subject…do the demographics of people who use the Internet differ a lot from the demographics of typical CHR listeners or the population in general? Are the people consistently responding to our surveys not likely to enjoy music styles that those without Internet access like? How much more reliable would callout research be? - Mike

Mike: Hmm…all things lead to the sample. Keep in mind that when you conduct any type of research using a sample of people from the population, you will have a certain amount of sampling error. This means the answers you received must be interpreted within the range of error related to the number of people involved in the research. In addition, all sample-based research contains random error and unknown error.

So you have a sampling error in your study. However, what complicates the problem is your statement that the respondents volunteer for the survey. What that means is that you can't compute a sampling error percentage—your sample is known as a "non-probability" sample because it is made up of volunteers.

It doesn't mean a whole lot that your samples are roughly the same size and age. What means a lot is that you don't know if these people are the same in reference to their perceptions of CHR music. I know you can filter things on a variety of elements when you run your analysis, but that doesn't cure the problem.

The problem with Internet research as you are using it is that you have no control over who gets into your sample. And that is not good. You don't know who is rating your songs. You say that the sample is the same, but you don't know. You don't know if these people are qualified. They could all be 55 year-old people.

If conducted properly, Internet research can be valid and reliable. However, if you're accepting answers from an unidentified sample, then nothing can insure that what you have is real. You are dealing with a crapshoot every time you receive your data.

You also ask if the demographics of people who use the Internet differ a lot from the demographics of typical CHR listeners or the population in general. In your case, the only way to know this is to conduct two tests…one on the Internet and one using callout or an auditorium test. You would then compare the results. (The tests would have to be conducted at the same time.)

Now, don't fall for the argument that your sample is large. That means nothing. You are using a sample of people who volunteer for the study. They are not randomly selected. The sample may be bad even if it contains 10,000 people.

As I mentioned, there is no problem with Internet research if it is conducted correctly. However, I haven't seen any music testing research that is conducted according to scientific principles. The research is usually conducted by non-researchers. And that's where the problems originate.

Is callout more reliable? That depends on how it's conducted.

Music Tempo & Gender Separation?

Dr. Wimmer: Do the listeners you have surveyed in the past indicate if they care about things like gender and tempo separation?  Some stations used to work hard to create a smooth "flow" of songs, slow to fast, and back again.  Today, most seem to just try to keep too many fast or too many slow songs from playing in a row.

Some stations used to say you couldn't play two females back to back, some even requiring a minimum amount of time or male artists between each female record. Today, the rule for most, but not all, stations seems to be never two females back to back, but one male is all the separation you need.

Related, as long as you are generally playing a variety of styles, do they care about style separation?  Thanks! - Gene

Gene:  In more than 30 years of conducting research for radio stations, I have never heard a listener complain or comment on a radio station's separation of tempo, gender, or variety.  These items are important to people who program radio stations, but they aren't important to "regular" listeners.

Now, someone might say that listeners don't know how to verbalize their likes or dislikes related to tempo, gender, or variety separation — that listeners "subconsciously" like the separation.  I don't buy that at all.  Most listeners are capable of describing what they like and don't like, but separations (of any kind) aren't on the list of important elements when they listen to a radio station.

I'm sure there are some people who are "die-hards" and stand by the separation rules regardless of who says what (or proves the point with research data), but there is a way to verify that average listeners don't care about separations of any kind.  The proof comes by analyzing what people listen to on their MP3 players (or iPods]).  Many (or most) people hit the shuffle button and listen to the songs in whatever order they are played.  There are no separations, but most people who have MP3 players seem to spend a lot of time listening to them even though there may be several male or female songs in a row, many songs with the same tempo, or even several songs with the same style.

Music Tests - National Charts Relevance to Net Radio

Hey Doc, I'm aware that a smart music director will only look at national weekly airplay charts as a barometer of a given track's popularity, and should give more credence to phone response and callouts/auditorium sessions.  However, since Net radio is not nearly as local/regional, are the national charts more relevant?  And, since the average Net station's audience (in general) is a mere fraction of an FM radio station, is it safer to rely more on the national charts—considering that an average Net station audience will provide less data (both quantitative and qualitative) on which to base programming decisions? Thanks. - Alden

Alden:  If you have been reading this column for a while, you probably know that I have said many times that I haven’t seen much use for the national music charts.  My criticisms include: (1) National data mean nothing for a local radio station.  At best, the data can only provide indications of what a local audience likes and dislikes, and the indications are very weak; (2) There is very little, if any, information about the samples used for the national charts.  The only thing usually mentioned is that the samples are large (whatever that means), and the insinuation is that the large sample makes the data “good.”  However, anyone with even a rudimentary knowledge of statistics and research knows that sample size means nothing in reference to the quality of the sample.  A sample of thousands can be worse than a sample of 100 if the large sample includes poor (the wrong) respondents.

Now…the development of Internet (Net) radio stations has provided at least some use for the national charts.  Most people who operate Net radio stations have no money for research, so we can’t expect them to conduct national music tests on their own.  The national music tests should at least provide some basic information about what listeners like and dislike.

However, my concern over the quality of the samples used still exists.  Before you use the data, make sure you understand who is included in the sample.  How are these people recruited?  Which screener questions are used to qualify/disqualify a respondent?  How many respondents are there in your target group?  If you don’t know anything about the sample, then you shouldn’t use the data—even for general indications.

Music: To Quantify or Not

I just got out a meeting where our GM wants to go to a "50 minute music hour" statement. We currently use "Half-hour of continuous . . ." Is there any research on the topic? Is it better to be ambiguous or to quantify? Thank for your help. As we close in on Thanksgiving, I'm thankful for your column. - Anonymous

Anon: Thanks for the kind comment. I appreciate that. I know you're not one of my kids because they don't read the column . . . "Why would I read stuff my dad writes? That's lame." Anyway, I'm glad you enjoy the column.

Since you have been reading the column, you'll probably guess that I'm going to tell you that you need to ask your listeners about the music quantification problem. You didn't mention why your GM wants to go to the 50-minute music hour. I assume, and know that I'm smiling while I type this, that he/she conducted a research project with your listeners and knows it's what they want. Yea sure, and monkeys are going to fly out of my….well, never mind.

I have asked the music quantification question many hundreds of times with listeners of all ages and in all formats. I have yet to find one group of listeners (this doesn't mean that your listeners will be the same) that prefers a number of minutes per hour quantification. Most radio listeners don't have any idea how much music can be played in 20 minutes, 30 minutes, or any other amount of minutes.

However, they do understand things like, "Another more music hour," or "More music all the time," or some other variation of this ambiguous approach. You are correct in saying that ambiguity is the best approach. It's best because it allows listeners to interpret the message in their own way. What does "Another more music hour" really mean? Heck, I don't know, but it sounds good to the listeners. In addition, when listeners are asked what they like about their favorite radio station, they say things like, "They play more music every hour." As compared to what? I don't know that either, but it works with the listeners and that's the most important thing.

When you get into the number of minutes per hour arena, you're entering what I refer to as "shadow boxing" because another radio station in your market will start using 51 minutes per hour, or 55 minutes per hour. Then what? You're stuck, because you can't play 60 minutes of music per hour. Your fighting your own shadow and the best thing to do is stay out of "counting" wars, similar to counting war of playing 10 songs in a row, or 12 songs in a row. I haven't found that counting wars are important to any listener in any format. They are important only to people in radio, such as your GM, because it seems like a good thing to say. Well…I say hogwash.

Since the early 1930s, persuasion theory has shown that ambiguous messages are almost always more successful than clear and distinct messages. The reason is what I just said…people can interpret the message in their own way.

Music "Too Black?"

Hi Doc, thanks for taking time out of your very busy day to answer our questions.

1.  Dovetailing on your recent Dance Music question: We are in a smaller market that has a rather low ethnic population.  Our PD routinely passes on songs because they are, in his words "too black."  Even though they can become tops of the airplay charts, he won't play them because they are "too black."  His pronouncements are the only "research" the station does.

2.  Do white listeners, on average, worry if a song is "too black" or is it more likely they like what they like, and tend to be "colorblind."  My PD insists that white audiences do not accept "black" songs or any rhythmic song.  He even has a rule that no two songs by black artists can play back to back.

3.  Is there any basis in truth for these "rules?"  Do people in smaller markets dislike rhythmic music?  Thanks. - Anonymous

Anon:  You're welcome for answering questions.  I'm glad you enjoy the column.

I numbered your paragraphs to make it easier to answer them.  My comments are probably going to come across as extremely discourteous to some readers, but frankly, these types of urban legends, myths (or whatever you call them) need to stop.  So here goes . . .

1.  I can't say that what your PD says is the dumbest thing I have ever heard about music.  What your PD says joins the others who have said the same thing, so all the comments collectively are the dumbest thing I have ever heard about music.  He joins the club.

But it really doesn't matter much about what I think because your listeners are the most important critics.  And I can tell you from about 20 years of conducting research on music that listeners (white listeners) would also agree that your PD's comments are uninformed, insensitive to listeners' likes and dislikes, and well, just plain dumb.  (I don't know if your PD will read this, but if he does, he needs to get his head out of the sand.  If his pronouncements, as you say, are the only "research" your radio station relies on, you radio station is in one gigantic heap of Vulpes Fulva droppings.)

With that said, I would really like to know what "too black" means in reference to music.  I don't know, and I have never heard one respondent in any research study ever use this term.  Too black?  Does that mean some songs are "too white?"  Are some songs "too Jewish?"  If I drank beer, I would go and get one right now.  Jeez.

2.  Colorblind?  As I just said, I have never heard one white person say that a song is "too black."  I have heard them say they hate a song, or love a song, but I have never heard one white listener rate (or judge in any way) music by the color or ethnicity of the person who recorded the song.  So yes, according to 20+ years of research with radio listeners, they are colorblind when it comes to who performs a song.

You say that your "PD insists that white audiences do not accept ‘black' songs or any rhythmic song.  He even has a rule that no two songs by black artists can play back to back."  In research terms, the first thing that comes to mind is, "Oh, please, gag me with a beaker."  I'm trying to think of an answer to your PD's statement, but I cannot.  I have never heard such a comment and it makes as much sense as not wanting to put two songs in a row by white artists, two by artists who can't carry a tune, or two Jewish artists.  Ludicrous, absolutely ludicrous.

3.  You ask if there is any basis in truth for these "rules" and if people in smaller markets dislike rhythmic music.  First, there are no such "rules."  The rules exist in your PD's head and I will not attempt to determine where his ideas came from.

Second, why would people in smaller markets dislike (or like) rhythmic music any more or less than people in medium or large markets?  That makes no sense, although I can say that it fits with the other things your PD believes in.  Oh wait, I forgot!  The difference must be that people in smaller markets only like non-rhythmic music because it helps them get through their day consisting of milking cows, shucking corn, castrating pigs, and sitting in the ol' outhouse.

In summary, I have never seen any research to support your PD's opinions.  He must be 92 years old or his first name is "Archie" (or both).

OK, my rudeness (justified in my opinion) is over.

Music: Ways to Play Songs

Let's say that I have 100 songs to play on my radio station. How many different ways can I play these songs? - AC

AC: Ready for a big number? In mathematical terms, what you're asking for is known as the factorial for 100, or the number of permutations (ways) the 100 songs can be arranged in distinct orders.

To compute a factorial (identified as n! in math terms), you multiply 100 x 99 x 98 x 97, and so on down to 1 . . .or 100!. I have no idea what the name of this number is, but here are the number of different ways you can play 100 songs . . .

93,326,215,443,944,200,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,

000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,

000,000,000,000,000,000,000,000,000,000,000

By the way, there are 141 zeros following "200."

Since there are so many combinations, and this is with only 100 songs, why do some listeners say that their favorite radio station seems to play the songs in the same order every day?

Music: Ways to Play Songs

Let's say that I have 100 songs to play on my radio station. How many different ways can I play these songs? - AC

AC: Ready for a big number? In mathematical terms, what you're asking for is known as the factorial for 100, or the number of permutations (ways) the 100 songs can be arranged in distinct orders.

To compute a factorial (identified as n! in math terms), you multiply 100 x 99 x 98 x 97, and so on down to 1 . . .or 100!. I have no idea what the name of this number is, but here are the number of different ways you can play 100 songs . . .

93,326,215,443,944,200,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,

000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,

000,000,000,000,000,000,000,000,000,000,000

By the way, there are 141 zeros following "200."

Since there are so many combinations, and this is with only 100 songs, why do some listeners say that their favorite radio station seems to play the songs in the same order every day?

Music Tests: Using Old Data

With budget cuts and a lack of needed music tests, what would you suggest to go back and look at in an old test, when competition has had time to react to your previous changes?  Burn?  Something other then total score to justify maybe playing more product that did not pass an old test?
 

Anon: It's soapbox time . . .

I understand the economy is in bad shape, and I understand that radio stations (and all businesses) are cutting expenses to survive.  However, what I don't understand is how radio station owners and managers can cut expenses for audience research and music testing.

Operating a radio station, like all consumer businesses, is a simple process: (1) Find out what the listeners (consumers) want; (2) Give it to them; and (3) Tell them that you gave it to them.  Now, it sounds as though your radio station followed this process in the past because you mention using old music test data.  That's great.  However, because of budget cuts, or some other reason, you no longer have current information and must resort to reviewing the old data.

That's a problem because all research, including music tests, has a limited "shelf-life"—the information is valid and reliable only for a limited time because things change quickly in radio.  Listeners change, your radio station changes, music (or talk) changes, your competitors change, and the environment changes.

I don't know when you conducted your last music test, but you need to determine if the information is still relevant.  If it's not too old (more than one year), then you can look at the data for indications.  Look at everything—familiarity, burn, comparisons between cells (male, female, ages, P1s, etc.)  You may find that instead of deciding to add more product, the only thing you need is to "tweak" the playlist.  Maybe a change in rotation would help.  I'm not sure.  You're the expert on your radio station, so determine if you are following the guidelines presented to you in your last music test.

Finally, the problem with management eliminating research and music tests is that the people who run the radio stations don't have current information to help them make decisions.  When that happens, listening often drops a bit (or a lot).  The next domino to fall in the process is that advertising sales drop because the Arbitron numbers aren't as good as they were in the past.  Smaller Arbitron numbers equal less ad revenue; less ad revenue means less money to invest in research and music tests.  Radio station owners and managers who cut research and music tests participate in a "self-fulfilling prophecy" for failure.  It's that simple.

Music IS THE PRODUCT on a music radio station, and it makes no sense to follow a course of action where decisions makers (usually PDs) don't have information about their product.

I'm sure many people involved in radio have heard managers, "We don't have the money for music tests or research (or several other things)."  If that's true, and I doubt it is, the radio station will have even less money in the future if research and music tests are postponed or canceled.

Muzak

What is Muzak? - Anonymous

Anon:  I’m sure there are many people who wonder about Muzak since many hear it everyday.

Basically, Muzak is a music service for any type of company that wants to play music for its customers (retail stores, bars, etc.).  The Muzakcompany started in the 1940s, and according to the Muzak website, the company has about 350,000 clients.  (About 70% of Muzak is owned by Abry Partners, LLC.)

I went to the website and it does explain a lot about the company, but I thought I’d go a step further.  I called one of my good friends, John Engelbrecht, who owns several radio stations (South Central Communications) and is also the largest independent Muzak franchise holder in the country, and asked him your question.  John has been involved with Muzak for several years, and he gave me permission to summarize his comments for this column.

Here is some of what John said….

Muzak for many years was known as the “elevator music” or “doctor’s office” music company because it provided only instrumental music.  In the early years, Muzak’s music was delivered to clients via telephone line—operators literally played 33 1/3 records over the phone line.  From there, Muzak used FM subcarriers to deliver its product, then changed to audiotapes that were shipped (bicycled) to customers.  Muzak now uses a sophisticated satellite delivery system.

John said that Muzak now offers about 60 different types of music.  Clients can program their music by broad daypart or literally any desired time frame, and they can receive several formats at the same time to play in different areas of their business.  Another advantage of the new Muzak is that clients can insert local messages.  Muzak is no longer an “elevator music” music company.  It offers virtually any type of music to fit a variety of clients.

If you’re interested, you can find out more information from this Internet search I set up for you.

All Content © 2012 - Wimmer Research   All Rights Reserved