PPM Insights

Pimping for PPM

Pimping for PPM addresses an important issue regarding Arbitron's roll-out of PPM. You can find our take on the matter at our sister bog Radio Insights here.

Arbitron reports third quarter revenues of $96.5 Million, up 6.4%

Arbitron Reports 2007 Third Quarter Financial Results

October 18, 2007 7:00 a.m.

                        Revenue up 6.4 percent 

Planned spending on Portable People Meter(TM) rollout drives increase in costs

Net income per share (diluted) is $0.58

NEW YORK--(BUSINESS WIRE)--October 18, 2007-- 

Arbitron Inc. (NYSE: ARB) today announced results for the third quarter ended September 30, 2007. For the third quarter of 2007, the Company reported revenue of $96.5 million, an increase of 6.4 percent over revenue of $90.7 million during the third quarter of 2006. Costs and expenses for the third quarter increased by 16.4 percent, from $56.8 million in 2006 to $66.1 million in 2007, due in part to planned expenditures for the rollout of the Portable People Meter (PPM) ratings service in Philadelphia, New York, Nassau-Suffolk, Middlesex-Somerset-Union, Los Angeles, Riverside, Chicago and San Francisco.

Earnings before interest and income tax expense (EBIT) for the quarter were $27.2 million, a decrease of 15.4 percent compared with EBIT of $32.1 million for the third quarter of 2006. Net income for the quarter was $17.2 million, compared with $20.2 million for the third quarter of 2006. Net income per share for the third quarter of 2007 was $0.58 per share (diluted), compared with $0.69 per share (diluted) during the comparable period last year. For the nine months ended September 30, 2007, revenue was $267.3 million, an increase of 6.9 percent over revenue of $250.0 million for the same period in 2006.

EBIT decreased 21.7 percent from $73.1 million in the first nine months of 2006 to $57.2 million in 2007. Net income for the period in 2007 decreased 20.2 percent to $36.5 million compared with $45.7 million in 2006. Earnings per share (diluted) for the nine months in 2007 were $1.21, compared with $1.51 per share (diluted) last year.

Stephen Morris, chairman, president and chief executive officer of Arbitron, made the following comments: "While the PPM commercialization is both complex and challenging, we have been able to stay on track with our ambitious market-by-market rollout schedule for the Portable People Meter ratings service. On September 20, we launched the 'pre-currency' survey period in New York and the embedded radio markets of Nassau-Suffolk and Middlesex-Somerset-Union. These three markets are scheduled to convert to Portable People Meter as full 'currency' on December 31."

"At the same time, we are recruiting consumers for Los Angeles, Riverside, Chicago, San Francisco and San Jose. While this has been logistically demanding, especially because each market has its own unique characteristics, we're committed to converting these markets as scheduled."

Arbitron Inc. (NYSE: ARB) is an international media and marketing information firm serving radio broadcasters, cable companies, advertisers, advertising agencies and out of home and online media advertising companies in the United States and Europe. Arbitron's core businesses are measuring network and local market radio audiences across the United States; surveying the retail, media and product patterns of local market consumers; and providing application software used for analyzing media audience and marketing information data. The Company has also developed the Portable People Meter system, a new technology for media and marketing research. Arbitron's marketing and business units are supported by its research and technology organization, located in Columbia, Maryland. Arbitron has approximately 2,100 employees; its executive offices are located in New York City.

The Company continues to expect that revenue will increase between 5.5 percent and 7.5 percent in 2007 compared to last year. Earnings per share (diluted) are expected to be between $1.35 and $1.45 for the full year 2007. These forward-looking statements involve known and unknown risks and uncertainties that may cause our results to be materially different from results implied in such forward-looking statements. These risks and uncertainties include, in no particular order, whether we will be able to:

      --  successfully implement the rollout of the Portable People
        Meter service;

   --  renew contracts with large customers as they expire;

   --  successfully execute our business strategies, including
        entering into potential acquisition, joint-venture or other
        material third-party agreements;

   --  effectively manage the impact of any further ownership shifts
        in the radio and advertising agency industries;

   --  respond to rapidly changing technological needs of our
        customer base, including creating new proprietary software
        systems and new customer products and services that meet these
        needs in a timely manner;

   --  successfully manage the impact on our business of any economic
        downturn generally and in the advertising market in
        particular;

   --  successfully manage the impact on costs of data collection due
        to lower respondent cooperation in surveys, privacy concerns,
        consumer trends, technology changes and/or government
        regulations;

   --  successfully develop and implement technology solutions to
        measure multi-media and advertising in an increasingly
        competitive environment; and

   --  successfully obtain and/or maintain Media Rating Council
        accreditation for our audience measurement services.

Orion, The Big Dipper, and PPM

PPM punditry reached full bloom at the NAB Convention. Session after session was devoted to sharing the “secrets” to winning in a PPM world. The confidence with which these pundits offered their advice far outweighed the evidence on which the advice was based. Of course, that has never stopped those of us in radio from telling others in radio that we know a secret to winning.

Most of the “secrets” had been gleaned by pouring over mountains of PPM data looking for patterns. The reasoning was that if the PPM numbers for a station spiked, there had to be a reason, and if we found the reason, we’d know what to do to create a spike in PPM. For example, a PPM time-line showed that the days WBEB, Philadelphia’s contest direct mail piece hit mailboxes, the station’s numbers spiked–and then declined on contest days. Does that prove that direct mail works, but contests don’t? The direct marketing people would like you to believe so. Arbitron would like you to believe so. It sells the value of direct mail, and it sells the ability of PPM to measure it. But does a PPM spike coinciding with one station’s direct mail piece prove anything? No, because other direct mail pieces have not had the same impact. Are contests ineffective? For one station they might appear that way, but there’s no way to be sure and other spikes on other stations have coincided with give-aways.

The problem is that humans by nature look for patterns. Early man looked at the stars and saw lions, hunters, and a scorpion in the pattern of stars. Today we know that constellations are nothing more than human imagination seeing patterns where none exists. As we see more PPM data, it is likely that we will again realize that many of the patterns pundits are seeing are nothing more than the PPM equivalent of constellations.
      
There is statistical “noise” in PPM just like there is statistical noise in every other method of audience measurement. PPM captures exposure, not listening. PPMs don’t capture every bit of listening. They miss chunks. PPM time stamps are reasonably accurate, but they can drift. People can carry their PPM one day and forget it another day. Editing rules can add quarter-hours in one hour, and subtract quarter-hours in another hour. All these things individually can add to the statistical noise of Arbitron PPM ratings. Collectively they can (and based on Philadelphia and Houston do) impact ratings by adding a degree of randomness to listening. Yet because we by nature seek patterns, we tend to overlook the randomness and focus on what we believe to be patterns.

The truth is, there is still very little hard evidence on what works for PPM measured radio stations. There’s a great deal of circumstantial evidence floating around, but proving that there is a correlation between PPM time-line spikes and any station action or event is very difficult, and proving that any action or event actually caused the spike is virtually impossible.

First, there is the issue of the number of panelists that can create a spike. In a typical demo, there might be as few as a dozen panelists contributing data. With a small base like this, a change on the part of just a few panelists can create the illusion of a significant change in listening patterns. Most of the “proof” that has been offered that something works has been based on very small numbers of listeners changing their behavior.

The second reason one should look skeptically at these claims is that the people involved have a vested interest in finding correlations. None of the studies presented by Arbitron would qualify as valid research in the academic world. The people involved in these projects were looking for spikes that coincided with something happening. That way they could “prove” the value and validity of PPM. That is not science, it is a sales pitch. Until someone produces real research, these puff pieces should be dismissed as marketing pieces.

As more markets switch over to PPM, we will begin to see patterns emerge. We will know they are real because they happen over and over to different stations in different markets. At that point it will be time to start paying attention to what the patterns are telling us. Until them, every time someone shares a PPM secret, look to the stars.

PPM's Tipping Point

The 2007 NAB Convention will be marked as a tipping point for Arbitron’s PPM. Four well attended sessions dealt with all aspects of radio’s new ratings “currency.” PPM now has the attention of broadcasters.

In contrast to previous Arbitron PPM presentations, this year’s were consistently upbeat and focused on the positive. Yes, we’re falling short in recruitment, but look what PPM can do! Yes, there may be only seven panelists listening to you at any given time, but we can tell exactly when one of them has tuned away! ( Well, actually because of editing rules, we’ll know more or less exactly when that person tuned away.) Inevitability was in the air at NAB. Ready or not, here comes PPM.