One of Arbitron's pitches regarding PPM was that switching to long term panels of PPM wearers would stabilize the numbers. We wouldn't see the dramatic swings from month to month. PPM would also enable us to look at more "granular" ratings. Monthlies would be more reliable, we could look at hour-by-hours with confidence, and we could even look at minute by minute ratings and make programming decisions based on what we found.
The graph below is the PPM share for 28 consecutive weeks of a single drive-time hour for the top rated station in its target demographic. The first week the station had about a 22 share, and by the end of the 28 weeks the station's share was down to a 9 share. More significantly, the station in week three achieved a 25 share only to plunge to a 7.5 share two weeks later. Imagine a station trying to make programming decisions on such swings.
Our look at stations in all three PPM markets shows a similar pattern. PPM numbers are no more stable than diary based numbers when we start looking at the "granular" numbers that Arbitron says PPM can deliver. Yes, PPM is an improvement over the diary method as long as we look at long time periods and large demographic cells. But don't expect PPM to give us the kind of "granularity" we have been promised.
Comments