"D Peter Maus" wrote in message
...
David Eduardo wrote:
"D Peter Maus" wrote in message
...
David Eduardo wrote:
"D Peter Maus" wrote in message
...
From a sample of the listeners. The rest of the process is
evaluation and interpretation by Consultants and PD's.

While some perceptual research requires interpretation, a music test
does not. It is simply play the top songs more than the next group, and
don't play the bad ones.

You've just made my point for me.

But there is no interpretation there. It is just play the number of good
songs needed, in ranking order, that you need to create the station
library. Unless a staition plays a lesser scoring song more than a top
scorer, there has been no interpretation. If a ball game ends with a 5 to
7 score, the one with 7 wins... similarly no interpretation.





Ok. Let's look at that.

How deep do you go on the list? You get one hundred songs rated, how
deep do you go? How do you tier them?

We already know that small variations are insignificant... just as we know
the Census was off by a rang of plus and minus 4 points.

So we can really look at the songs based on belonging to sets. All those in
a particular set have, statistically, the same score within the margin of
error. Each set is a tier. Each rotates in proportion to popularity.

Every format has a close range, in any market, of songs that are more
positive than negative. So your basis is songs that are positive without
significant negatives. You get your playlist without looking at artist and
title.

How do you rotate them.

In proportion to popularity.

With what do you mix them?

Other songs of different levels of hittiness.

How old do you go?

As far as the listeners tell you.

What's your current/gold ratio?

In proportion to the currents vs. golds that test.

How many categories of Recurrent do you have?

Depends how many recurrents test. Many staitons change categories and
rotations based on what tests... if there are fewer recuerrents, then the
category(ies) are smaller.

Do you play the top 10 straight up, or do you break it up into top 10 and
top 5 like WLS did in the 70's.

Based, here, on weekly call out. There are seldom more than 3 very hot
songs, and they rotate faster if they exist. If there are no standouts, then
they rotate alike.

With different rotational periods for every category.

Based on scores and the math of rotation. Like a big swiss watch full of
gears.

There you have four decisions, interpretations, to be made by
Manglement that separates a list of songs from Music Programming.

And all four are really determined by the characteristics of the songs that
tested.

Your sample picks songs.

Your PD (Local or Regional) and his/her Consultant programs the music.

There are very few consulted stations these days, and the ones that do exist
do not program the music... they help the PD develop and implement
programming strategy... and maybe to implement the test. The PDs, except for
the untestable new songs, only play what the listeners picked in proportion
to scores.