A Must-Read...Part III in a series of articles from Steve Rivers with Todd Wallace.

Steve Rivers: A few weeks back, the MusicBiz crew visited some record labels in New York. One of the hot button topics that came up was callout. How are programmers using it? Are they using it as excuse for not adding a song because it doesn?t test well? The frustration, I think, comes not from callout itself, but the fact that programmers throw it up as an excuse for songs they?ve placed in callout after only a small number of spins on their station. How do you respond to that?

Todd Wallace: Well, it is true, callout is most effective in measuring a product that has established a reasonable pattern of familiarity. The closest we've ever come to addressing this, is a function of our MARS system called, "Hit Potential" which analyzes a song's favoriteness in relationship to its familiarity and projects a score that can call a programmers attention to a song that's starting off strong. It really is very accurate in spotting a hit that?s fast out of the chutes. But, that doesn't apply to all songs. Some songs are first listen hits that wow you from the beginning, yet other songs are slow to burn in and very slow to burn out. So, here again, a programmer needs to apply a dose of common sense and interpret results with a grain of salt.

One of my pet peeves as a researcher is the PD who institutes an arbitrary score threshold before a song receives airplay or has been played in any significant rotation.

They'll say, "Well, after 3 weeks, it's only achieved a PosAcc reading of 39.2%, and my cutoff line is at 40%." How short-sighted! Meanwhile, he'll keep on playing some midchart stiff that's been stalled at 43% for 9 or 10 weeks, but which meets his silly over 40% or nothing rule. Some PD's just can't quite see that there's something wrong with that picture. One of the ways you can illuminate the interpretation process is to start testing a song even before you?re playing it. That way you can establish an early track on its Q-score progress to see if it?s gaining in Hit Potential. Also, if your competitor is playing the song and you aren't, you might check the other station's cross tab numbers instead of the total line, since it's going to be more familiar to their P1s than to yours.

SR: As a programmer interested in generating the highest ratings possible, I would not only use callout to put together my song rotations each week, but I would also tabulate requests, look at sales and seek the commonality with all of these to put together the strongest power rotations possible. I'd look at it from different camera angles, as you would say.

TW: I love getting different camera angles on everything! The trick is in knowing just how much weight to attach to each angle when you interpret the overall big picture. Frankly, I wouldn't put requests or record sales on a par with a properly executed callout program, but I would certainly take note of any breakouts from either requests or sales. If a song is Top 10 in either sales or requests, you?d better be testing it and probably playing it.

SR: My experience is that at any given time there are no more than three to five true smashes, and that the rest is filler for the most part. I used several layers of recurrents and some limited power gold, but with the secondary currents and the news, I think you must try and focus on using songs you feel have the potential to become powers, and that fit the sound of your station. Are you finding that it takes longer these days for a song to surface as a hit than say, 10 years ago? If so, why?

TW: I like the way you think. And I agree with your figures. Actually, I've found that it usually is three bona fide hits and three or four almost hits. I always used to tell my Top 40 clients, "Any record we play is, was, or will be a hit." Actually, I guess that applies to just about any hit axis format. A soft AC station plays the hits for 25-54s or 35-64s, a Country station plays the hits for that life group, and so on.

The quick answer to your question is, yes, I think it does take a little longer for a hit to make it, which is largely the result of the fragmented marketplace you find in most Top 50 or even Top 100 markets these days. And with these lines of format demarcation, we don't see as many multi-format crossovers as we used to. That having been said, however, I would submit that if you super-focus on only your station?s super-core, you'll probably find that the hit-process life-cycle is very similar to what it was ten years ago. One reason it might be a bit slower is that the older an adult gets, the less important music is in their lives. They become less inclined to get excited about a new song. For example, a Soft AC station might find true hits slower in developing than, say, a younger targeted Triple A or Hot AC would.

SR: How long would you say is enough time to give a song before giving up on it?

TW: I don't think you can put an arbitrary time window on exposure of a new song. There are so many variables. I think you would be shortsighted in just looking at number of weeks.

I think you need to take on board various factors. Mostly, I think it's a function of the number of cumulative spins you?ve given a song or perhaps the cumulative spins in your market and your format, especially if it's a song with crossover potential. You should also take into consideration your station's efficiency in converting cume into quarter-hour listens, and how that will impact on an OES plan of Optimum Effective Scheduling.

You can track that just like you would the effectiveness of a commercial campaign. Then there's one of my favorite old adages: Some of the very best radio today is still gut-feel radio. The corollary to that, of course, is: It's just how well informed you keep your gut.