THE TOP 10 INSIGHTS

From the Advertising Research Foundation Annual Convention

New York, March 5-7, 2001

1. Census Bureau wins the Ogilvy Award as the year’s most outstanding advertising developed through research. Y&R partnered with a host of ethnic agencies to produce advertising in 17 languages encouraging minorities to return their census forms. They were so successful it has been decided no further efforts to adjust Census results for under-reporting by minorities will be needed.

2. "The Tipping Point" is a new key to researching new products. Malcolm Gladwell, author of the popular new book by that name, addressed the convention and described it as the point where a few well-informed "Mavens" and a few gregarious "Connectors" can start a new idea spreading like an epidemic. A morning was spent discussing the research implications. Rosi Ware, until recently President of Millward Brown and currently a consultant to the Target Group Index (TGI), described how TGI’s semi-annual single source omnibus surveys in 30 countries validated Tipping Point Theory. Psycographics defined early adopters that correctly predicted new product successes in ice cream, finance and wine. Anke Audenaert described Yahoo!’s Buzz Index as another new tool based on the same idea. Yahoo! makes daily counts of the number of times your brand, product category, or process, is the subject of a search query by Yahoo!’s 166 million monthly users. The first paper on new products was from IRI’s Valarie Scalla. She mined IRI’s scanner database for 600 new products and came up with a number of key findings: More new products in a category means more failures, more advertising helps, and only one in four have test marketed the new product. 80% failed to get past the $10 million threshold for sales in the first year. Another new IRI study on mergers showed most brands falter after acquisition.

3. Use of many types of advertising is a key to "Surround Marketing". One of the first issues raised about this latest addition to the marketing vocabulary is also one raised in our BRC talk (See item 4). How do you measure success when you are using a multiplicity of media? The answer for Olive Garden was increased sales. That‘s fine for the advertiser, but not as satisfying for the researcher. There was nothing to show the gains were related to the advertising. Without ad tracking, or other evidence, there was nothing to disprove the possibility they could have achieved the same gains without spending anything on advertising.

4. Recognition-based tracking should replace GRPs. Paul Shellenberg and I suggested this. GRPs and the whole system of ratings-based media analysis has always had trouble showing how many are being reached by a firm’s advertising, or how often, because the amount of overlap between media is largely unknown. Now the situation is getting worse because TiVo, and Replay are making it easier to skip commercials. Channel surfing during ever lengthening breaks also reduces the commercial audience without changing the reported program audience. Then, there is the proliferation of channels and magazine titles. Recognition-based tracking was suggested as a solution because it only counts advertising respondents actually saw. It does this much more accurately than recall, and it can show if an individual has been reached by TV, print, radio, web sites or any other media. So, it provides precise measures of reach and overlap. Likability and diagnostics show if people were affected by the advertising that reached them. Studies offered as validation of these procedures were the ARF Validity Study (1991) that showed likability and diagnostics were the best measures of advertising impact, far better than recall, persuasion and main point feedback. This kicked up enough tumult that 40 of the country’s largest advertisers kicked in to fund IRI’s "How Advertising Works" (1995), in part to check the ARF results. They confirmed the findings, concluding both recall and persuasion were poor measures of advertising’s impact.

5. Ipsos–ASI recommends recall and persuasion measures for all pretesting. They brought this up twice, first in describing their pretesting of direct to consumer (DTC) Rx advertising, and then in describing Next*Idea, an early stage screening process to find ideas worth developing further. So much for any consensus on the best way to measure advertising’s impact.

6. Hershey reported recall of ads not aired for years. Kirk Ward, their Research VP, also reported he has never seen a heavy-up study, where GRPs are increased, producing a significant increase in sales. Modeling the results has shown the contribution of advertising copy quality is critical. Modeling has shown companies are past the point of diminishing returns on advertising and should cut back. In- store promotion will also help to offset the effect of price.

7. Fusion offers another way to handle the unknown overlap between media. A whole morning was devoted to its discussion by folks who are into media research, indicating the importance placed on the topic. Fusion refers to combining databases. A database showing what TV programs people are watching could be "fused" with the results of another survey showing people’s use of other media. It’s based on a key assumption. If the people in the TV database had been asked about their use of other media, their answers would have been identical to those from the same type of people in the other media survey. Thus, a new database is created that could show how many viewers of Program X are likely to read magazine Y and listen to Station Z, combinations neither database could have shown by themselves. Fusion could be used in the same way to generate numbers showing how many viewers of Program X buy Fords, use Crest or eat at MacDonald’s. The best ways of doing this, and testing the accuracy of the results, were discussed and debated. But all seemed to agree some form of fusion, long used in Europe, would be adopted here. It was felt syndicated single source research, that measured everything needed in a single survey on a timely basis, was never going to be cost effective. There was relatively little discussion of how well this idea, that has been under consideration for over a decade, would be received by other researchers, and the advertising community in general.

8. Nielsen agrees at the last minute to help study fusion. Leslie Wood, Chair of the ARF’s Media Methods Committee, proposed an industry fusion laboratory to test the possibilities from a statistical and numeric perspective, as well as a more pragmatic view: does it help make better decisions? She showed an impressive group of participants had already signed up, including the producers of the two databases where successful fusion would have the greatest payout: MRI and Nielsen. Both agreed to provide their respondent level data for experimentation. What she diplomatically didn’t mention was Nielsen’s agreement had just been received the day before, on this item that had been on the agenda for months.

10. There was concern Yahoo! was over committed. They just joined the ARF. They announced their new Buzz Index (Item 2) at the convention. The Convention Chairman, Murray Gaylord, was from Yahoo!. The company president was a scheduled luncheon speaker, to be replaced last minute by the sharp and funny Jerry Shereshewsky, Direct Marketing VP at Yahoo! He described how Yahoo! could hardly avoid being a great opportunity for direct marketing. They know who has been searching for what. All this, the same week Yahoo! made headlines announcing lower earnings, and their stock went into the dumper. It left cynics buzzing. Was their new interest in research a cause or an effect?

10.5 Lots of other insights: Online surveys still show bias toward "geeks", but not as much as before. (NFO) David Haig of the UK consultancy Brand Finance reviewed the financial implications of brand equity and came up with an eye-opening conclusion: Market Research should not report to Marketing. That’s not where the decisions are made. They should report directly to Finance or Management. Keynoter Peter Dolan, CEO, Bristol-Myers Squibb, said it’s now common for patients to walk in and ask their doctor for prescription drugs by name. AT&T’s heavy use of the bouncing, squirming globe logo was inspired by the Actors Guild Strike. (AT&T) Video Research from Japan uses the percent staying with the brand as a measure of the long term effect of advertising, and the ratio of the number who have started considering the brand to the number who have stopped considering it as a measure of short term effects. DiscoverWhy can now collect consumer’s responses continuously as they watch commercials online and superimpose them over a copy of the commercial.

These are only highlights of things that caught my interest. For more complete details check the ARF website at www.theARF.org for the availability of papers, or call the ARF at (212) 751-5656.

Don Bruzzone, March 2001

Bruzzone Research Company · 2515 Santa Clara Avenue · Alameda, CA 94501- 4692 · (510) 523-5505 www.Bruzzone-Research.com