American association of advertising agencies

August 10, 2006

VIA EMAIL

Ms. Susan Whiting

President and CEO

Nielsen Media Research

770 Broadway

New York, NY 10003

Dear Ms. Whiting:

On behalf of the members of the Media Policy, National TV/Radio, Local TV/Radio, Media Research and Media Technology Committees of the AAAA's, the following sets forth a summary of consensus points on some of the important issues surrounding Nielsen's proposed release of the "average commercial minute rating" tapes. We offer these insights so that you may have an opportunity to discuss them with the Committees and their members in the coming weeks and months.

In summary, the Committees request the following:

• For the all-minute tape, Nielsen should continue to provide flagging of individual minutes that contain commercials and strive to provide flagging of individual seconds that contain commercials.
  
  

• Any data released by Nielsen needs MRC accreditation.
  
  

• Any syndicated product that calculates an "average commercial rating" should remove VCR recording from the equation, until Nielsen can provide VCR playback.
  
  

• Nielsen should evaluate the various industry-proposed calculations of an average commercial minute, and compare their statistical reliability to one another and to the current average telecast rating.
  
  

• Nielsen should advise the industry how it plans to distinguish between national and local commercials for cable, and include cable networks not currently measured by Monitor Plus. National and local commercial activity for syndication should also be discussed and evaluated.
  
  

• If Monitor Plus is to become the backbone of commercial audience estimates to be used as currency, the industry needs to be ensured of its validity. It will need MRC audit and accreditation for all national television activity (network, cable, syndication). What are Nielsen's plans in this regard?
  
  
  
  

Page 2

S. Whiting

August 10, 2006

Research and Negotiation

The Committees believe there are clear lines between research issues and matters of negotiation between buyers and sellers.

• Whether to use Live, Live + Same Day, or Live Plus 7 Day is a matter for negotiation between buyer and seller. From a research standpoint, there is no significant difference in terms of statistical reliability among the various streams of data. Nielsen should not take sides or even appear to take sides on this issue.
  
  

• Upfront buys for the next broadcast year have been based on Live data. That is the currency of the moment. Nielsen therefore has an obligation to the industry to provide (on both tape and hard copy reports) audience data that corresponds to the industry's current trading currency. If any ratings report or computer system can only handle one set of data, it must correspond to the currency of the moment.
  
  

The potential use of any average commercial minute rating will eventually become a matter for discussion between buyers and sellers. For now, however, it is a research issue requiring extensive evaluation by Nielsen and subsequent discussion through industry committee.

• Nielsen's assertion that it will release multiple commercial minute rating tapes based on various client requests is quite disturbing. It is one thing for an individual client to request a custom product for analytical purposes for internal use only. But when Nielsen offers to make multiple custom-requested products available to everyone (at a cost) for "use as potential currency", it threatens to thrust the entire industry into chaos.
  
  

• We believe Nielsen has a responsibility to the industry beyond simply being a data provider. When offering data as a syndicated industry product, the level of scrutiny must rise beyond what is necessary for an individual client's request. When Nielsen offers a syndicated product to the industry as a whole, Nielsen is implicitly (and explicitly) standing behind the data and vouching for its validity from a research standpoint. This is the main reason Nielsen delayed the release of its weighted national sample until convinced that it met its research standards and rose to the level of currency.
  
  

Several research issues are in play regarding an average commercial minute rating.

• How should we define a "commercial minute"?
  
  

• There may be significant differences in the accuracy of the data when comparing different methods of calculating an average rating. This should not become a matter of negotiations. It must be driven by the highest quality research as determined by Nielsen (or industry evaluation facilitated by Nielsen).
  
  

Page 3

S. Whiting

August 10, 2006

A disturbing response came out of the recent Nielsen Client meeting in New York. When questioned about the validity and stability of the proposed data that the broadcast networks were requesting, Nielsen's response was that it was up to Nielsen's customers, not Nielsen, to decide if the data was stable enough to use as currency. We believe that Nielsen has the responsibility of doing the testing and giving us the statistics that will enable an intelligent analysis. There must be a very high research standard for any measurement tool to potentially be used as trading currency.

• One agency has done research that indicates that for primetime broadcast, the typical 60-minute show has 10-12 individual minutes that contain at least 30 seconds of national commercial time. Does averaging 11 individual minutes provide as stable and robust a rating as averaging 60 minutes? These are not the same 11 minutes in different episodes of the same program. How does this differ for shorter-duration programs? How does this affect the stability and projectability of the data by demographic segment? What are the corresponding minute counts for cable, syndication, and other broadcast dayparts? How does the stability and projectability of the data over time compare to current average telecast ratings?
  
  

• VCR recording should be removed from the ratings for any calculation of an average commercial minute. The network argument that since Nielsen can't measure VCR playback, it is equally unfair to remove recording from the ratings as leaving it in, is understandable when it comes to program-based ratings. When we're talking about commercial measurement, however, this argument no longer holds up. The move toward commercial type measurement provides the opportunity to correct this situation. We believe Nielsen has an obligation to provide the best research possible, and do not find it helpful for any average commercial minute tape to be released containing VCR recording, unless playback can be accurately measured by you.
  
  

The AAAA supports Nielsen in its efforts to improve research via its A2/M2 plans, and to provide additional types of research data for analytical purposes. As has been the case in past situations, however, a higher level of confidence is required when addressing anything as fundamental as potentially changing our industry's trading currency. Research issues must be resolved through careful consideration, not because one segment of Nielsen's client base can agree on what they want.

We offer these thoughts in a spirit of cooperation and constructive dialog. We look forward to your response, as well as future meetings to address the issues raised by the AAAA, the CAB, the broadcast networks, and others affected by this situation.

Sincerely,

Judy Vogel

Research Director, OMD

Chair, AAAA Media Research Committee

cc: AAAA Media Committees: Media Policy, National TV/Radio, Local TV/Radio,

Media Research, Media Technology

AAAA: Donna Campbell, Claire Franks, Mike Donahue

Nielsen: Sara Erichson, Jack Oken, Dave Thomas, Catherine Herkovic, Kevin Svenningsen,

Pat McDonough, Paul Donato