PROGRAMMERS STILL VOICING CONCERNS ABOUT PPM DATA All Access On the heels of the most well-attended Arbitron Consultant Fly-In in its history, and extensive facts and presentations from Arbitron designed to add to radio's knowledge base of the new ratings gathering technology, there are a number of programmers who still have some unanswered questions, and who have a desire to share their questions at-large, in an effort to help radio have a clearer understanding of the PPM and its formulas.
Among those on record are Randy Kabrich Consulting Pres. Randy Kabrich who also represents a number of Cox Radio outlets, and veteran radio programmer and Bill Tanner & Associates President Bill Tanner. Kabrich is credited with going public with the fact that Houston PPM carry rates and in-tab were off 33% due to a technical snafu from Arbitron. Two days later, Arbitron announced the glitch had been fixed.
Both Kabrich and Tanner have shared their thoughts with All Access and other broadcasters at-large over their latest PPM concerns. Arbitron SVP/Press and Investor Relations, Thom Mocarsky, has been contacted for comment.
Tanner, who speaks up publicly for the first time, told All Access, "I have seen the Arbitron charts and heard the slick 'data glitch' denials. I watched as they used 'rating points' to demonstrate PPM stability; easy to say because rating points are so rounded. I've heard 'Rankers don't matter that much.' Try telling that to buyers angling for a lower rate. Try telling that to stations in a fierce, neck-and-neck battle. Let the PD say that to the manager or owner. How many programmers get bonuses based on Cume?
"It's here that the lack of PPM respondents matters. Delivering 55% of the 18-34 target in Philadelphia matters. Weighting matters. It matters that Arbitron doesn't disclose the ethnic in-tabs by age/sex, as they promised to do on last week's call but have not. Why not? Surely it's right there in their computers. Why don't they disclose the 'per person meter value', as they did with 'per person diary values'?
"Indeed ratings are compressing, and better sampling may well confirm that. But better sampling is what the radio industry is paying for and, as Arbitron's Adam Gluck said, bigger samples mean greater reliability."
Tanner goes on to take issue with tools Arbitron has provided programmers. "The DMR study presented Thursday afternoon (8/23) pointed out the huge importance of P-1 and P-2 listeners to a station's ratings. I believe that is true. And, where is that in the PPM Analysis Tool, which is software so slow that radio users have to set it, go to lunch and then come back for their report, only to discover the computer has timed out?"
Tanner continues, and has asked hypothetically ... "Regarding Houston, which is it:
A: No one at Arbitron noticed until after the reports had been released? Or:
B: Arbitron did see it but just decided, 'Losing in excess of 200 weekly respondents doesn't matter. We'll weight it up. A minor data glitch.'
"Another question:
A: Arbitron never saw any 'seasonality issues' or 55% in-tabs of the 18-34 demo issues in the years of pre-testing? Or:
B: We did see it but we released PPM anyway. Weighting will take care of it."
In closing, Tanner said: "Arbitron thinks it's the government. But, in fact, it is a highly profitable monopoly, doing very well financially, increasing bills for PPM 65% or more. They are field-testing PPM on the backs of an industry that is struggling."
Kabrich Says: Hmmmmm!
Consultant Randy Kabrich, who has been very vocal, and has lead the charge in dissecting Arbitron's PPM numbers, told All Access, "When Arbitron finally released the Houston corrected August Week 1 in-tab numbers, it showed 1,189 In-Tab Daily and 880 In-Tab Weekly.
"According to Arbitron Spokesman, Thom Mocarsky, and quoted in the Chicago Tribune, 'We forgot to tabulate 200 or so people" -- which explains why the weekly in-tab number went from 683 to 880 in the correction.'
"But when one takes 880 (weekly) / 1,189 (daily) this shows that one would expect a 74% rate in daily translating into weekly in-tab. However, these 200 or so respondents they 'forgot to process' seem to have a 99% rate in daily to weekly.
"So the 200 or so respondents they 'forgot to process' had a 25% higher rate than normal (and one would add, a rate unseen in the history of PPM). Hmmmmm.
"Of course, Mocarsky also was quoted in the Chicago Tribune as saying 'the Houston incident was a one-day glitch,' a simple data-processing error that it corrected the next day."
Kabrich takes issue with Arbitron's Quality and Control procedures, and wondered aloud why the data wasn't updated and released until last Thursday, taking exception to Arbitron's claim that this glitch makes no difference in the Daily Numbers. "Obviously, the weekly number has no so called direct impact on the Daily AQH and Daily Cume rate. But...if the person had undocked and used the meter every day (6 or 7 days of the week) of course it would have had an impact on the Daily AQH."
Speaking to Arbitron's acknowledged under performance in Houston, in August Week 1, Kabrich quipped, "If a radio station just happened to leave out 20% of their monthly payment to Arbitron, much like Arbitron did in the Houston PPM sample, do you think that the accountants at Arbitron would just let that slide? If not, then perhaps Arbitron needs to move some of their accountants to Quality Control."
Kabrich minced no words, throwing down the gauntlet for Arbitron to perform to promises, and offers a temporary solution. "While PPM is clearly the de facto winner, Arbitron has yet to proven they can deliver what they have promised the Industry.
"Arbitron is releasing the PPM results for New York at 1PM on December 31st, just so they can count the revenue in this year's revenue for Wall Street and push their revenue higher. Why doesn't Arbitron do something for Radio -- and delay all markets for 6 months until they can prove they can deliver Houston and Philadelphia as they have promised?" |