So What’s Hot?

by Richard Harker

“What are the hot formats?” That’s a question we are frequently asked. And while it seems like a simple question, it turns out that determining which formats are growing, and which formats are declining is not an easy question with a simple answer. The answer depends on how one tabulates the results. If in fact, the best known reports on the health of formats turn out to be all wrong. In many instances, formats can appear to be hot even while their popularity is collapsing.

For many years, Arbitron has produced the American Radio Listening Format Trend Report. Available on their web site, the trend report extends back to 1998 and lists 15 broad formats subdivided into a total of 60 different formats. Arbitron generates a format share by combining Average Quarter-Hour Persons (AQH) of all the measured stations throughout the US. So the shares represent the percentage of Americans in measured markets who listen to a format. For example, in Spring of 2006, CHR had 11.3 shares, in Spring 2007 CHR had 10.9, and way back in 2002, the format had a 12.0. One might naturally assume that this is a format losing listeners. And it is, as Arbitron calculates shares.

The problem is that the number of stations in a format affect the total quarter-hours. As stations change format, each is registered as a loss of AQH for the old format and a gain for the new format. This presents a distorted picture of what is happening within both formats. If stations abandon a format, the total number of quarter-hours for the format decline. With this approach, Arbitron may report a decline in the format’s share even if the stations that remain in the format have not lost listeners. At the same time, stations switching to a new format can give the illusion of a format that is growing, just because new stations are adding quarter-hours to the format. A second problem with this approach is that stations with high AQH have a greater influence in the totals than stations with smaller AQH. This can result in a few highly successful stations having a disproportionate influence on the apparent strength of a format.

In defense of Arbitron, the company makes no assertion about the predictability of their Format Trend Report. They simply publish the latest ratings. It is the pundits and the press that tend to read more into the report than there is. When the results are reported, many just assume that whatever has happened will continue. So if Arbitron reports growth in a format, there is a tendency to assume that it will continue. Unfortunately, history has shown just the opposite. Formats that appear to be growing in the Format Trend Report are likely near their peak and are likely to decline in the future.

For many years Katz Media Group has calculated format trends somewhat differently.  KMG calculates a share for each of 25 formats by averaging the shares of stations within each format in those markets where the format existed. Format performances extending back to 1990 can be found at www.katz-media.com. While Arbitron’s approach gives greater weight to stations with large AQH, the KMG approach tends to give greater weight to medium to smaller market stations. That’s because stations in smaller markets generally have larger shares than stations in the top markets. As a result, the results under-weight large market stations. This produces larger shares with the KMG method. For example, in the Fall of 2006, Arbitron reported a total share of 14.8 for Adult Contemporary while at the same time KMG reported a total share of 20.2, a not insignificant difference.

A somewhat simpler approach has been taken for many years by M Street. Their format trending report is a tally of the total number of stations in each of 31 formats. The tally available on-line is monthly and extends back to September 2006, although data back to the early 1990s is available in print. Compared to Arbitron’s approach, the M Street tally is more equitable–it gives equal weight to every station regardless of market size. However, this approach is particularly susceptible to a problem we call the bandwagon effect.

The number of stations in any format rises and falls based on its popularity within the industry. For example, the number of Country stations has varied from a high of 2,613 in the mid 1990s to today’s low of 2,046. One might assume that with a 20% drop in the number of stations doing Country, that the format must be in trouble. Quite the contrary, Country is doing fine for the stations in the format. What has happened is that many marginal stations have switched from Country to more trendy formats in hopes of better ratings. This is the bandwagon effect.

One such trendy format was the 80s format. Initial success in markets like Chicago and Seattle caught the attention of broadcasters and soon there were over 20 stations doing 80s. Station counts and total quarter-hours for the format grew very quickly, suggesting that the format was on fire. However, beneath the frothy headlines, the format was in trouble. The stations that had been in the format the longest were losing listeners even while other stations were switching to 80s.

Because of the bandwagon effect, all three traditional methods tend to act as lagging indicators. They generally lag what is currently going on with formats. It is said that when a social trend appears on the cover of Time magazine, it is already over. We’ve observed a similar phenomenon with format analyses. They do a pretty good job of telling us what has happened, but they do a poor job of predicting what will happen in the future.  More often than not they indicate a format is headed in one direction when it is actually headed in the opposite direction. AQH growth and station counts can both be on the rise while stations within a format are suffering declines. These indicators can also be suggesting that a format is in decline when the majority of stations are growing, but weak stations are bailing out of the format. Last year when many large market Oldies stations were dropping the format, and the trades were predicting future troubles for Oldies, the majority of Oldies stations were actually growing.

Because all these traditional methods of identifying hot formats suffered from serious flaws that created false impressions about the health of formats, Harker Research set out in 2000 to find a more meaningful way of trending formats. After studying the flaws and weaknesses of the current methods, we determined the best method to see how formats were performing was to see how stations already in a format fared. We found that by eliminating stations new to a format and looking at only those stations that were already in a format, we could eliminate the impact of the bandwagon effect.

In 2001 we began tracking all stations in the top 50 markets. We identified the format of each station and then sorted the stations into one of 22 separate broad format categories. For the purpose of determining the strength of each format, we first eliminated stations that had changed format during the time period. We also eliminated stations below 1.0 share because of the volatility of shares of stations below a one share. We then tabulated the proportion of stations that increased their share from book to book and then expressed the number as a percentage of stations in the format.

For six years now we have tracked these formats. Our method has proved to be the first leading indicator of formats. It signaled a cyclical up-trend for Country even while stations were abandoning the format and Arbitron’s Format Trend Report showed a decline in share. Our approach signaled a turn-around for CHR even while national shares and station counts were trending down.

So what’s hot? Comparing the Summer Book results to the Spring, the big music winners were Alternative, Urban AC, and Classic Hit stations including the Jack stations. News, Talk, and Sports all did well. The big losers this Summer were Christian formats. After years of steady growth, the format seems to have peaked a few books ago and is trending down. Other losers this Summer include Active Rock, Oldies, Smooth Jazz, and Movin.

While book to book comparisons are the most common, our research suggests that the most reliable indicator of a format’s health is looking at year to year comparisons, in other words, comparing Summer 2007 to Summer 2006. Looking at the Summer book this way, we get a somewhat different (and we believe more accurate) picture of what is happening within formats. The big winners in the year to year derby are Adult Contemporary, News/Talk and Oldies.

Christian formats were down both in book to book and year to year comparisons.

For the Fall book analysis we will be able to include the ppm markets. Share figures are scheduled to be released to the trades as well as weekly cume. Our measure of format health will be a reliable transition as we enter into the new ppm world. If you have any questions or comments, please email us at info@harkerresearch.com.

Richard Harker is President of Harker Research, a company providing a wide range of research services to radio stations in North America and Europe. Twenty-years of research experience combined with Richard’s 15 years as a programmer and general manager helps Harker Research provide practical actionable solutions to ratings problems. Visit www.harkerresearch or contact Richard at (919) 954-8300.