Q & A: Where Do Radio Station Ratings Come From?

I was asked a question the other day about radio station ratings and surveys and where it all comes from… Information

So today, a primer on how radio and other media get this information and what it means.

There are two independent companies that conduct surveys to determine who listens to radio stations and watches tv shows.

I’ll focus on radio, you can ask someone in the tv business how it works in their business.  While it is similar, there are some differences.

The two companies that measure radio listenership are Eastlan and Nielsen.  Nielsen bought Arbitron which was the standard for decades.

However Arbitron/Nielsen has had some flaws in their methodology that they were slow to fix.  Along comes a new research firm in 1999 designed to overcome the complaints that were not being addressed by Arbitron.

Arbitron used a diary method of record keeping. Listeners were contacted by phone to qualify to be in a survey.  Then they were mailed a survey book to fill out for 7 days and return by mail.  The flaws in this system were multiple with the changes in technology.  First off you had to have a landline phone, cellphones were not included.  With the number of homes discontinuing home phones, this became a problem in getting a big enough sample especially of certain demographics.  Then just because you qualified someone and sent them a diary to fill out, would they do it?  The response rate shrunk the sample even more.

Accuracy of the data in the diary was always questionable. But Arbitron tried to overcome all of this with a new way of collecting data. …Arbitron’s launch of the Personal People Meter, or PPM, in 2007. The PPM is a small, wearable device that measures what radio stations people listen to.

However the PPM with its hand off approach is expensive and also considered inaccurate. Only Major and Large Markets are using PPM’s.  Fort Wayne is ranked as market #113 and Nielsen uses the old paper dairy method.

Along comes Eastlan and they have a plan to build a better research company.   From their website:

  1. Sample Size Eastlan sample sizes are significantly larger-often by hundreds. Eastlan also believes in surveying only one person per household for broader representation.
  2. Weighting No response rate problems here. Eastlan’s hybrid of random stratified telephone-recall sample and daily e-surveys is very precise and closely matches the population composition of the survey area. By contrast, diary methodology surveys tend to be heavily weighted as telephone calls or direct mail are used to place diaries but there is no control over which diaries are actually returned.

Now you know a bit about how the data is gathered.  By an independent research company (Eastlan) that sells access to their research to radio stations and others who want to know.  When I mention that WOWO has over 100,000 listeners, that is from the information I have access to from Eastlan.

I can drill down into the data, using the software from Eastlan and pull reports and data on all kinds of demographics for Fort Wayne radio stations and their listeners.

Some of the information is interesting but not very useful.  For example I can see that WOWO averages over 7000 listeners that are age 18 and older at any time during the week, compared to the next highest station which has less than 5000.  Here’s why this is not very useful.  This number represents an average number of all 168 hours in a week, 24/7.  A more useful number would be the total number of weekly adult listeners 24/7.  Again WOWO dominates was the only station with over 100,000 adults age 18 and older. The #2 station has 10,000 less according to the rating data.

Surveys in Fort Wayne are done twice a year, in the Spring and Fall and we get the results a few weeks after they are concluded.

Before I wrap this up, I’ll touch on the other advertising media options and how they measure their audience.

Print which includes magazines and newspapers offer counts in a couple of ways.  There are the number of subscribers and the number of papers published.  Neither number really tells us how many readers they have.

Billboards use traffic counts.  Using data from the highway departments they have estimates of how many vehicles drive past each sign.  Again, this doesn’t really count how many people looked up and read your message as they drove by.

How about the web and digital marketing?  We have gotten better at determining how many people have seen a page, an ad or visited a website and clicked on something.  However there are still several missing links in the chain of events that leads from that first exposure online to tracking the purchases.  I learned this when I worked for an internet based retailer and saw differences in our own internal tracking systems and Google Analytics.

Questions?  Go ahead and ask.