Barrett Media produces over 20 stories per day on the music, news, and sports media industries. Stay updated on the latest happenings by signing up to receive our newsletters straight to your inbox.
If you’re a regular reader, you know that my preferred method for analyzing PPM data has always been the PPM Analysis Tool (AT) which, as I’ve noted before, is the second-best export from Portugal after aged port (Taylor Fladgate 20-year-old is wonderful and has a fairly reasonable price). In this column, I’ll cover some other AT features, specifically how granular you can go with the data for a radio station.
Spend some time with Analysis Tool and you’ll find that you can create almost any demographic group you want if there is enough sample (30 or more) to run it. If you click on “Target” when you do a run, you can always find the standard demos easily, such as persons 25-54 or women 18-34. But if you click on the “Custom” tab in the “Target” option, you’ll find many different ways to lay out a target.
Start with age and you’ll see that you can run any age to any age. Would you like to see 7-14? I wouldn’t know why, but if there is enough sample, you can do it. 31 to 47? No sweat. If you’re interested in the Hispanic audience, you can break it down five different ways with respect to language capabilities and seven different ways for country of origin, again assuming enough sample.
Some other options include the less-known weighting variables, employment status and presence of children, along with household size, household income level, education, and more. A caveat: household income level in Nielsen Audio World still tops out at $75,000+ which is essentially useless in most markets. The 2023 data from the American Community Survey (conducted by the Census Bureau) showed that the median household income in the Boston metro area was over $110,000, San Francisco-Oakland-San Jose exceeded $127,000, and the Washington, DC metro was $121,000. Remember, the median is the midpoint meaning half are above and half are below; it’s not an average.
Also, be wary of education levels. This is probably the prime variable for respondents being less than truthful, not just to Nielsen, but in general. Typically, high school dropouts do not declare that they have Ph.D.s, but rather they move themselves up one level. High school dropouts will say they graduated from high school. Those who didn’t finish college may claim that they graduated. During my Arbitron days, I remember our demographer, Dan Estersohn, gave me an analysis of diary service data showing the high school dropout rate being far lower than what the Census Bureau claimed.
People do this on resumés as well. When I was in the process of being hired at Arbitron in 1999, the headhunter verified my claimed degrees. I wondered why, but her comment was something to the effect of “you’d be surprised how many people lie about their educational achievements.”
AT can do other party tricks with granularity. You can look at unique dayparts and unique measurement periods. When I did analyses for Cumulus stations, I would average the individual days of the week, for example a quarter’s worth of Mondays, Tuesdays, Wednesdays, etc., and compare the listening levels for each day of the week. One hint: you should remove any major holidays because the levels are typically lower on those days and will compromise your averages. Just click on the “Survey” button, then choose “Custom” and “Add”. A calendar should pop up allowing you to create whatever measurement period you want.
The result was often that a particular day of the week had higher levels. With PPM, it won’t be an artifact of the measurement system as in the diary service, where Thursday is always the heaviest listening day because it’s the first day of the survey. If you find a difference in a key demo, move your biggest promotions to that day when more people can listen. You’ll also want to check on your and your competitors’ stations’ strongest and weakest days of the week.
Everyone in the business is way too busy today and with the possible exception of yours truly, no one has a great deal of time to experiment with what AT can do. But try it…if it means that you’ll understand your cluster’s audiences or your direct competitors’ audiences better, taking advantage of AT’s flexibility and granularity will be worth it.
Let’s meet again next week.
