Make Every AMT Song Selection Better

A client recently asked me about the selection of songs to include in his upcoming AMT. He had more songs than he could squeeze into a single test.

Among his concerns was whether he should make more room by dropping some of the titles that "always test well."

Here are my thoughts about selecting songs to test:

Rule one: Always test the big songs! I believe that you should be trying to put together the very best test list possible.

Those big songs don't really always test ... at least not all of them, and not with every demo. Plus, there is always the need to monitor them for burn.

If you use a research company that, like Steve Casey Research, does music clustering and fit analysis, leaving out key songs that anchor and define your format will cripple your analysis.

I suggest that the 100 songs that tested the worst in the previous AMT, or up to 200 songs (if there are that many) that tested badly in both of the last two AMTs, can be left out. Substitute different songs that you haven't yet tested but have reason to believe will test in the top half of the test.

Tip: If your research gives you some kind of "fit" information, leave in the odd song that tests very bad but has a decent fit score against Pure Core. That kind of song has a better chance of making a leap in the next test.

Be sure to look at the most compatible music clusters for good clues as to the types of music that you should be mining for additional test songs. Don't just toss in a song because you play it as spice on the weekends. Believe, for some reason, that it has a shot at the top half of the test.

If you don't have any kind of cluster breakdown of your music, you'll have to use your own good judgment alone. But either way, look for more music of the type and from the artists that are already working. That rule may not always apply to currents, but it is very reliable when you're looking for more oldies.

Not Only Listeners Should Attend Your Music Test

Another way to increase the value of your music test is to get participation from people who are responsible for the success of your radio station. Here are some ideas from 2003 that are just as valid today.

AMTs are unique. They put you face to face with a truly representative cross-section of your listeners. Your people should take advantage of that.

Does your General Manager/Managing Director attend your AMTs? Does your Program Director?

Sometimes when I attend a music test, I notice that none of the above people stop by. Unusual? Unfortunately, no. Maybe I'm too sensitive. But I think this business is all about a passion for entertaining and serving. The audience was right there! People with a passion for radio shouldn't be able to resist an opportunity to get close to and observe the people.

Are your listeners nothing more than statistics to be manipulated? Are they just numbers that (hopefully) get bigger? If so, maybe running a grocery store would be a bit more to your liking?

There is something else you should have your people do. They should actually participate in the AMT.

• They will learn if it is easy to understand the task.
• They will learn if it is easy to express their opinions in the time allowed.
• They will learn whether it is fun to participate in an AMT.
• They will learn, just a little, what it is like to be a listener instead of an employee.

A typical story from a recent test: While checking answer sheets during the break, I noticed that hook 141 was getting consistently high ratings. One of the station's people who was taking the test had, however, indicated that she was "tired of" this song. Interesting, so we scurried around to find the list of titles and check on this mystery. The song was "(Everything I Do) I Do It For You" by Bryan Adams.

And what a typical result! A song that the listeners were crazy about -- while station people are tired of. It is this result that is at the heart of why we must do music research, and do it well. We who are in the business do not think like "normal" listeners. We embrace music with greater passion and we tire of it more quickly.

Get your people more involved with your music testing. You'll be glad you did.