- November 4, 2008

Whither the Click?

Impact of Online Display Advertising Absent the Click
Part I: Its Impact on Site Visitation

Last month, the IAB invited me to present a summary of Comscore's Q2 2008 e-commerce trends as part of an IAB webinar where the IAB released their Q2 online ad spending data.

The IAB data were assembled and presented by Price Waterhouse Coopers (PWC) and were most interesting. Overall, the data were reassuring, with a growth in online ad spending of 13% versus year ago. Search continued to grow strongly (+24% vs. YA) while display ads (i.e. banners) grew by 8%. Consistent with the PWC data, CMR also reported an 8% increase for display ads. Let’s hope we see more of the same in Q3 and Q4!

At about the same time, Nielsen released its own estimate of display ad spending for the first half of 2008 and reported a 6% decline. I was intrigued as to why the Nielsen number differed so widely from the PWC and CMR estimates and spent a little time digging into the numbers. It turns out that Nielsen reports online ad spending for CPM-based display ads, while the PWC and CMR numbers include both CPM and “pay-for-performance” display ads (i.e. CPM, CPC and CPA deals).

So, in interpreting the data, it would appear that advertisers are shifting large amounts of their display ad spending from CPM to CPC / CPA deals. With today’s tough economy, I don’t think this is surprising – since it’s natural for advertisers to demand more performance. But, I think there’s more to the story. In a study Comscore completed earlier this year with Starcom and Tacoda, we measured click rates across all online display campaigns in a month and found them to average less than 0.1%. Does this mean that display ads aren’t having any impact? I don’t think so. I think the issue is that a click no longer reflects the effectiveness of a display ad. Just as we wouldn’t expect that print ads, TV ads and radio ads should generate immediate consumer response, why would we expect it to be so with online display ads?

We've conducted extensive research at Comscore that confirms the impact of display ads. Using our behavioral panel, our analysis compared the behavior of consumers who were exposed to display ad campaigns with a control group who were not exposed. The control group was carefully selected to be demographically and behaviorally balanced with respect to the exposed group. Below, I’ve shown the average impact of 139 display ad campaigns in terms of their ability to drive visitation to the advertisers’ sites:

The impact of the display ad campaigns is clear, with substantial lift in site visitation occurring in the first week of exposure to the campaign (+65%) and continuing through the fourth week following initial exposure (cume +46%).

So, with these types of positive results, are we to conclude that advertisers who pay on the basis of CPC arrangements are making a mistake? I think the answer is “yes and no.” “Yes,” because it’s clear that a singular focus on display clicks is misleading and does not reflect the actual impact of the ad campaign. "No," because smart advertisers understand the benefit of running display campaigns and that paying based on the number of clicks realized is economically very attractive. They get the “view through” impact of the ads but only pay for the small number of clicks that are generated. That’s a great deal for the advertiser -- but a poor one for the publisher. I think it behooves publishers to consistently measure “view through” impacts and to use the results to charge a fair price for the holistic performance of display campaigns that are run on their properties.

In my next blog posting I’ll reveal further evidence of the effectiveness of display advertising by focusing on its impact on consumers’ search queries.

More About

Ad Impressions