• by David L. Smith , May 6, 2002

Sometimes it seems like the Web is trying to replace readers per copy with copy per reader and it won’t work.

My hope is that once I write this column, some major vendor will come forward and say, “we’ve got the data that you are looking for.”

But the fact remains that I see no good Interactive industry data on the potential degradation of recall and awareness as a function of ad clutter on commercial websites. The reality is that for even the most vanilla portal or major site pages, there are at least 3-4 ads. Call them what you will, banners (at the top of the page), sponsorships (buttons), a small tile of half banner in some “unused” space, skyscrapers (in that “unused” column to the right in larger monitors/more granular display sizes), throw in a house ad or two, some text link -- they are all ads. And that seems to be the minimum. Toss in a few pop-ups or pop-unders, add these to a site that has a lot of “strategic alliances” (translate that to multiple buttons and links on the page in addition to all of the above elements) and you get pages with 10+ and sometimes 20+ advertising elements, especially if you add the paid placement sections (sometimes obvious, a la Overture, but sometimes not transparent at all to the user, a la Amazon).

In traditional media we know these relationships. A break in a TV program averages 70% of the value of in-program. Why? Because the break between programs is all clutter, sometimes with 10+ ads before you get to any more content. And the commercial break in prime only has two minutes or four :30 spots.

We also know that a :30 scores 70% of a :60 and a :10 scores at least 50% of a :30 in recall. The same is true for print. We know the value differential of a half page vs. a full page and how this differentiates for magazines vs. newspapers.

Now that branding is accepted on the Web (go with me on this), the next step is to determine how to deploy our impressions most effectively. We know in other media that all impressions are not alike. We understand that the first impression has disproportionate influence. And that after 10, they don’t mean as much any more. We also know that if consumers leave the room, they are not much good at all.

So, how are you taking into account the premium that an advertising unit alone on a page, or with 2-3 other different kinds of units is valued vs. the unit that has many other similar units on the page or is one of 10+ advertising units (whether they are text or graphics does not matter here, the consumer will figure it out eventually)?

So, who’s got the data? Put it out to the industry. Let’s start developing the effectiveness models that we can apply to various sites based on their willingness to feature our advertising in a foreground mode, rather than jam it in with a ton of other advertising. And then, let’s be willing to pay a premium in branding advertising when sites will feature our advertising in a foreground mode. Sure, session-based advertising is one answer, but there have to be others. We want to know what happens when we run smaller vs. bigger ads. And what happens when we are one of four ads on the page vs. exclusive, or one of 15. And, once we get the data, we are going to reflect this in our RFP’s and negotiations.

Planners need it to sell it.

And buyers need it to evaluate it.

And, but the way, we need this real soon now.