Yahoo Consortium moves into second phase

The new phase of the Yahoo ConsortiumA major effort is underway by Yahoo and members of the newspaper consortium to promote the value of the behavioral targeting (BT) software that Yahoo is providing newspaper partners. Early results apparently show increased click-through rates and some happy advertisers, although we don’t know the actual numbers yet. The consortium of 800 newspapers includes many of the nation’s largest, and they have bet their online business future on the value of being able to serve local banner ads to Yahoo users in their markets. The revenue is shared with Yahoo. The BT software also allows them to target Yahoo users who visit their own properties, so the concept is potentially attractive to both the partners and Yahoo.

In a Sunday New York Times article, three papers owned by E. W. Scripps were featured, The Knoxville News Sentinel, The Ventura County Star and The Naples Daily News. Each reported big revenue numbers from sales “blitzes” with local advertisers, so they’ve been able to generate excitement in the communities they serve. There’s a lot at stake, too. Scripps has already closed one of its flagship papers, The Rocky Mountain News.

Ken Doctor, a newspaper industry analyst with Outsell, told The Times that papers are counting on this partnership working. “For the companies that are in it,” Doctor said, “this is the No. 1 growth initiative in 2009 and 2010.” The Newspaper Association of America meets next week in Las Vegas, and we expect to hear much more about the behavioral targeting aspects of the consortium there.

But some observers, including usability guru Jakob Nielsen, strongly believe that the long-term viability of any kind of banner strategy is a losing proposition. In an email about The Times article, Nielsen told me that the problem isn’t targeting; it’s banners.

Jakob NielsenTargeted is probably *slightly* better than non-targeted, but not by much. The main problem is that users ignore the banners. When you never see a banner, it doesn’t matter whether it’s targeted or not.

As a separate question, it’s not clear how quickly users’ intent and interest decay after they have conducted a search. In the moment of a search, we know that users are fairly likely to click on an ad that’s advertising the thing they searched for, as indicated by a keyword match between the query and the advertiser’s specification of the campaign. But how useful is the same keyword match for predicting users’ willingness to click 10 minutes later, let alone 10 days later? Particularly in a context where the user is no longer actively looking for that item, but rather browsing for something completely different. The answers to these questions are not known, and I am not aware of any independent research to look into them. (By independent, I mean: not sponsored by somebody who’s trying to sell these ads. Because if you have an agenda, you can make a research study show almost anything by setting up the parameters “right.”)

Ultimately, though, the decay rate is less important than the basic fact of banner blindness. Yes, if users’ intent decays slowly, then targeted ads will perform a bit better than if users’ intent decays fast. But in either case, the predominant factor will be that the users don’t even see the banners, meaning that they won’t click them whether or not they would hypothetically have been interested in the offer.

Gordon BorrellGordon Borrell’s company reported the decline in banner advertising efficacy last year, and he told me last week in Dallas that he will closely examine the new click-through data but doesn’t expect major changes in their position. Banner advertising on already cluttered media company sites does little to move the rock for advertisers, and that’s the problem. People may “see” local ads on, for example, Yahoo email pages — and they may generate leads and sales — but the evidence that local advertisers don’t need expensive media company ads cannot be ignored.

This is why we feel so strongly that local media companies desperately need to be moving resources to the creation of new value for themselves, value that is not associated with ad-supported content models (See: Flat is the new up below).

The Yahoo Consortium is a very, very important deal for everybody involved, and I’m not trying to throw cold water on the possibilities it offers. Targeted advertising IS an important part of our Local Ad Network concept, and the Yahoo software may play a role in that downstream. Banner ads may be more visible on non-media sites, but we don’t have any research on that, because local ad networks simply don’t exist yet. That will change, however, so stay tuned.

Trackbacks

  1. […] Yahoo Consortium moves into second phase | PoMo Blog — From Terry Heaton: [S]ome observers, including usability guru Jakob Nielsen, strongly believe that the long-term viability of any kind of banner strategy is a losing proposition. In an email about The Times article, Nielsen told me that the problem isn’t targeting; it’s banners. […]

Speak Your Mind

*

This site uses Akismet to reduce spam. Learn how your comment data is processed.