What is research?

[MightyFrenchman]

Research is an excuse for management-level people to try to avoid responsibility for their station's poor performance in the questionable ratings game. It's also an excuse to avoid creativity in programming. Radio "research" is as reliable as my 1978 Chevette wasn't.

 

[neutralobserver]

Research is an excuse for management-level people to try to avoid responsibility for their station's poor performance in the questionable ratings game.

Oh, gosh, no, it's not. It's a tool a station can use to avoid poor performance altogether!

It's also an excuse to avoid creativity in programming.

Oh, gosh, no, it's not. It's a tool a station can use to build a framework for creativity in programming!

Radio "research" is as reliable as my 1978 Chevette wasn't.

Oh, gosh, no, it's...???...come again?

What is "research?" "Research" is finding a statistically-significant, -representative, and -reliable sample of existing consumers and/or target consumers and asking them what they think, feel, like, and want.

You know what? These kinds of comments come only from people who've never used or been able to use it. (Once you use it, you come up with different whiny complaints. Well, no, not really...I was just trying to be funny.)

 

[ChiefEngineer]

Research is a useful tool for those who can adapt or who can take direction.

Every business from radio to McDonalds can benefit from it.

 

[jackd]

Research is an excuse for management-level people to try to avoid responsibility for their station's poor performance in the questionable ratings game.

Oh, gosh, no, it's not. It's a tool a station can use to avoid poor performance altogether!

It's also an excuse to avoid creativity in programming.

Oh, gosh, no, it's not. It's a tool a station can use to build a framework for creativity in programming!

Radio "research" is as reliable as my 1978 Chevette wasn't.

Oh, gosh, no, it's...???...come again?

What is "research?" "Research" is finding a statistically-significant, -representative, and -reliable sample of existing consumers and/or target consumers and asking them what they think, feel, like, and want.

You know what? These kinds of comments come only from people who've never used or been able to use it. (Once you use it, you come up with different whiny complaints. Well, no, not really...I was just trying to be funny.)

I agree 100%. I will also add by saying that you can't make your decisions TOTALLY based on the research. Good radio programmers know how to mix the statistics with gut. But, bottom line...allow your listeners to give input by using research & then LISTEN to their opinions. When someone complains about a song, I simply ask if they are on our Music Advisory Team. ;D

 

[MightyFrenchman]

I'd like to join your Music Advisory Team. I want to hear music by Randy Matthews, Andrae Crouch, and Rick Riso. Also toss in some Pat Terry Group.

[/quote]
I agree 100%. I will also add by saying that you can't make your decisions TOTALLY based on the research. Good radio programmers know how to mix the statistics with gut. But, bottom line...allow your listeners to give input by using research & then LISTEN to their opinions. When someone complains about a song, I simply ask if they are on our Music Advisory Team. ;D
[/quote]

 

[CONSultant]

Yes. Research is only a TOOL and should only be used as such. It is only as good as it's methodology of set up. Where stations get into trouble is following the research as the be all and end all. How were the 4 to 8 second clips of music used in your auditorium test chosen? How can the auditorium decide a song is great if they never hear it in the test? Do you have to blindly following the results of a test done in another part of the country for your station? How high is your burn rate set and why?

Research in radio is just like research in other industries; if it's not conducted and used properly it can give very skewed "results."

 

[neutralobserver]

I will also add by saying that you can't make your decisions TOTALLY based on the research. Good radio programmers know how to mix the statistics with gut.

...and I think that most often comes into play when an intelligent programmer spots "prestige bias" in research results.

"Prestige bias," for those not familiar with the concept, is the idea that research respondents themselves add bias (the generic definition, not necessarily the connotative definition that many think of when they hear "bias") to a survey's results by "discerning" the "best" response and giving it even when it's not true of them...in other words, answering a question the way the respondent thinks it "should" be answered even if it's the exact opposite of the respondent's true opinion.

Sticking with the Christian realm, let's say you absolutely hate "Amazing Grace"...you never liked it, its melody is too simple, the lyrics are loaded with words nobody uses, and so on. Yet, when someone brings it up and asks you how much you like it (a biased question to start with, by the way), you think you would look really bad and bring disapproval to yourself if you said what you really think about this long-loved standard and, so, say you really like it. That's "prestige bias."

Or let's say someone asks you what you think about Nicole C Mullen's music. You think, "gee, if I say I don't like it, does that make me a racist?" and then answer with that in mind. That, too, is "prestige bias."

So....an intelligent researcher and user of research is going to be aware of the possibility of prestige bias and have a good idea where it might show up, making that one example of a good and valid use of "the gut" in conjunction with research.

However, one should be aware that getting a result the exact opposite of what one's "gut" is saying does not mean the research was bad!

 

[GaryTheThompson]

Good, thoughts on research.

I will add that online music testing is usually not able to give a truly accurate result based on the fact that you can't REALLY environmentally or demographically control the cross section of respondents.

Auditorium testing is easier to control and make truly scientific.

If a station is relying upon an online 'music team' or some such, they're setting themselves up for a fall.

Some say "well, flawed research is better than no research."

Is it?

 

[gr8oldies]

OK, so let's do like everybody on these boards seems to say, throw out the whole idea of research, after all, we know what our listeners want more than they could possibly know. In fact, if I just play all of my favorite songs, I'll be number one. We have no need to ask our listeners..OK, maybe the ones who show up at remotes for the free hot dogs. And we of course need to play every conceivable genre of music all at once, have a 10,000,...no, wait, 20,000 song playlist (so what if somebody has to wait months to hear a song they like again. Of course, when you can't make payroll you won't want to research why.

 

[smashedcd]

research is some CONsultant in a suit telling you what to play instead of you deciding for yourself

 

[GaryTheThompson]

Research is absolutely necessary if you want to win in most cases.

The only situtation where I can see research not being an absolute necessity is if you're the only game in town with heritage and a track record of successful programming behind you.

But...the product could always be better even if you're the only game in town and you're succeeding. So why not use research?

Just seems like common sense to me. Why wouldn't someone use all the tools at their disposal to be as good as they can be? Isn't that just good stewardship?

 

[smashedcd]

to neutralobserver..

i saw the post that got erased. i have been outside of this market. i lived in atlanta. is that market big enough for you? i have spent time in nashville. i still like wsm-am. and i worked at 3 radio stations bc (before clear channel)..you may not like me. and you may call me an idiot. i really dont care what you call me actually ;D i am about 40. and as i stated i use to love radio. i just want fm radio to be fun again. but i know it never will be. at least not in most markets.

oh i was once on way fm daily..(1999- early 2000).


smashed
remembering the good old days of
102 zzq
98 rock wcko
z106 wstz (bc)
96 rock atlanta (bc)
wls


oh well i still have xm ! and sirius!

 

[radioelizabeth]

research is some CONsultant in a suit telling you what to play instead of you deciding for yourself

try again.

research asks the LISTENERS.

And any PD not listening to the listeners and just spinning what they have solely decided for themselves has no business stating they are attempting to reach a target outside the mirror.

 

[neutralobserver]

Careful, Elizabeth.....telling an enthusiast the truth is liable to get your posts moved to another section. :

Better lie to him, instead. Go ahead and tell him something like, "no matter what you hear, the only person researched is the program director's love interest."

Or, "consultants tell local stations exactly what to play and say so that they can hear stations they personally like as they drive all over the country."

Or, "consultants, though not on the org chart or even on the payroll, have supreme dictator power."

Or, "consultants use secret methods to find out exactly what message board posters love...and then recommend the exact opposite just for spite!"

 

[JimmyJames]

The idea that what a potential audience of hundreds or thousands can be determined by short clips of songs makes little sense. I never decide on something before hearing it through completely, and it takes multiple spins before people become aware of a song, therefore testing them on a few seconds of it with just a small group of people is not an accurate way to program a radio station. Testing is letting the impulse of a minority decide what goes on the air, and that's a dangerous precedent.

 

[GaryTheThompson]

I disagree.

Those stations who use testing (properly and scientifically) see better results overall.

To put it in the words of Ron Burgundy: "It's Science..."

 

[JimmyJames]

Why would I base the music decisions on my station on someone's opinion of a few seconds of a song? I want their opinion of the whole track, and to do that, first I decide if it fits the sound and tastes of my audience and then play it. Yes, we don't hit every time but when we do, we've done so by giving all our audience a chance to hear and decide for themselves, rather than have a select group keep that track off the air by an impulse feeling driven by a short clip. How can anyone realistically evaluate a song without hearing the entire song at least once?

 

[MightyFrenchman]

Who is Ron Burgundy? Is Ron the one who said, "It's science when you spit in the wind and divide by 12"?

JimmyJames, you're absolutely correct. The most desireable audience/demo groups that a radio station wants to reach aren't going to participate in an "auditorium" song test.

I disagree.

Those stations who use testing (properly and scientifically) see better results overall.

To put it in the words of Ron Burgundy: "It's Science..."

 

[neutralobserver]

The idea that what a potential audience of hundreds or thousands can be determined by short clips of songs makes little sense.

Actually, it makes a lot of sense and works very, very well.

I never decide on something before hearing it through completely,...

Not the 50th time you're hearing it, you don't. By then, you already know it and have a like or dislike. You recognize and react.

...and it takes multiple spins before people become aware of a song, therefore testing them on a few seconds of it with just a small group of people is not an accurate way to program a radio station.

...which is why hooks of unfamiliar songs are not tested! "Hooks" are the most familiar, recognizable, and catchy pieces of the songs; by definition, an unfamiliar song does not have a "hook." (Well, it does, but it's all but pointless to try anything with it.) With new, unfamiliar songs, the entire song is offered.

Testing is letting the impulse of a minority decide what goes on the air....

No, not at all. It's taking or arranging a statistically-valid sample of actual and/or desired users of the product and using a time-tested method to find out what they want and enjoy most.

But, not stopping there, if you think about it, you refuted your own argument -- is there any minority smaller than that of the radio station's program director??! Any decent music testing is going to be 6000-400,000% larger than that!

 

[neutralobserver]

Why would I base the music decisions on my station on someone's opinion of a few seconds of a song?

Because a few seconds is all it takes!

If you have someone who you think has a lot of potential as a programmer but who does not have an appropriate view of music research, you can do the following and create a convert pretty much every single time. Assign that person the task of preparing a hook tape from scratch. (That's a bit of a dated thing to do, but stay with me.) Monitor him or her to make sure he doesn't allow the hooks to get any longer than seven or eight or so seconds. Odds are he's going to think they need to be longer and object; keep 'em short, anyway. Then, assuming that that person would not be recognized by those participating, take that person to an auditorium test and put him or her right in the room as his own hook tape is being played. A funny thing will happen within five songs -- the doubter will see that the respondents begin recording their responses within about one second of hearing the hooks begin! Well, golly gee...the listeners know the music! Imagine that! The doubter's doubts vanish.

I want their opinion of the whole track, and to do that, first I decide if it fits the sound and tastes of my audience and then play it.

Why you? Who are you? They're them...the ones who actually use the product! You're just you.

Yes, we don't hit every time...

Why take that risk and blow it so many times when you don't have to??

...but when we do, we've done so by giving all our audience a chance to hear and decide for themselves...

...and we still can and do. Now, though, we can do that off-the-air. Isn't it better to do it off-the-air?

...rather than have a select group...

It's not a "select" (as in "elite") group. Depending on the research format chosen, it's a randomly-selected or volunteer group.

...keep that track off the air by an impulse feeling driven by a short clip.

That's an interesting paradigm -- you believe listeners keep songs off the air. I believe listeners put songs on the air! See the difference? It's a big one!

How can anyone realistically evaluate a song without hearing the entire song at least once?

The same way you can realistically evaluate a bite of food the moment you put it in your mouth -- your senses and brain work very, very quickly!