I blogged about this subject here at ReveNews.com about a year ago btw.

The user simply seeing an Ad a number of times at the right time and right frequency has impact on the conversion down the line at direct response campaigns. Studies have shown that too many or too frequent exposures also can have negative impact on conversion further down in the funnel.

People except certain kind of advertisers to advertise at certain places in order to demonstrate their status within an industry. People see the ads and might not react on them (e.g. click or even convert), but if the ads were not there at all, the person who is ready to buy might not considers that company as much as he would have, if the appearance of an advertisement by that company at a major portal in the niche wouldn’t have reinforced the status of the advertiser as one of the leaders in the space.

The same psychology is going on when it comes to buying paid ad positions in the search results, even if you are already ranking no.1 organically for a certain keyword, including your brand.

A great read. I believe (supporting data anyone?) that you also see this banner impression effect just with PPC where a merchant’s PPC ad appears at the top, and helps to support CTR/CR for an affiliate PPC ad appearing lower in the auction (especially for smaller brands).

A study by Atlas is hardly unbiased. As part of what was the Aquantive family and now part of Microsoft Advertising, Atlas provides the tracking and optimizing for a large portion of the display advertising online. It only behooves them to publish a study that demonstrates the tie between search and the work which is core to their business model.

But let’s use Atlas as an example. Back when DrivePM was also part of the Aquantive family I did several large media buys of remnant display advertising on a CPA basis (technically it was a hybrid model but let’s not get too caught up in the minutia). DrivePM ran those remnant buys using Atlas’s tracking system to optimize banners based on how they converted. Ones that didn’t convert, Atlas automatically optimized out of the system.

In this case, unlike typical display or remnant campaigns, the publisher, the tracking network and the advertiser were all in sync to insure the best possible conversion was reached while at the same time creating brand awareness and benefiting search. If I had purchased the same amount of display advertising directly solely on a CPM basis I would not have achieved the same value in part because the publisher (much less the tracking network) has NO incentive to optimize the banners it displays on my behalf.

PS. Your basketball analogy is well done.

Thanks for the comments

I never disputed the fact that you should also measure display ads individually. In fact, it would be ridiculous not to.

Also, I don’t think I mentioned that display advertising “performs well” by itself. I said it performs well at driving conversions for other online channels such as search. As I mentioned above, “While display ads might not always convert well on their own, they do perform very well at driving conversions, traffic, and brand awareness to other channels of online advertising such as search.”

My comments aren’t simply based on my data. It’s based on data from many other studies such as the one from Atlas Solutions that found, “When marketers supplement search with display impressions, they get a significant lift in conversions. Unfortunately, most advertisers that run both search and display are unaware of this…” The study demonstrated that “users exposed to both search and display ads convert at a higher rate: an average of 22 percent better than search alone and 400 percent better than display only.”

http://www.atlassolutions.com/uploadedFiles/Atlas/Atlas_Institute/Engagement_Mapping/eMapping-TP.pdf

The problem is that most web analytics solutions don’t provide attribution metrics across multiple campaigns. So an advertiser would never know, for example, that his display ads are valuable at driving conversions for his search ads.

Metrics such as Assists in Yahoo (or any other campaign attriution metrics) are very valuable for advertisers when it comes to making smart online advertising budgeting decisions. Think about it as a sports analogy….

Let’s say you were a basketball coach and you had to make a decision about which players you were going to play and all you had were conversions to look at. Let’s say in a game that Player A scores 2 conversions (for 4 points) and Player B scores 7 conversions (for 14 points). If you only had conversions to look at, which player are you going to bench in the next game based soley on conversions? Player A of course because of poorer performance.

But let’s factor in Assists now. Let’s say that while Player A only had 2 conversions (for 4 points) he did have 12 assists (which helped his team score 24 points). Now are you going to bench Player A? No, because while Player A didn’t convert well himself, he was very valuable at assisting his team in scoring 24 points (12 Assists).

Imagine the consequences of benching Player A. You would have lost out and additional conversions as a team because Player A wasn’t around to provide assists.

Same goes for managing the budget for online campaigns (display, search, email, etc). Imagine an advertiser lowing the bid for one of his search keywords because it didnt have a sufficient amount of conversions atached to it. But if he was using the “right metrics” (like Assists from Yahoo for example), he would have recognized that the keyword he just lowered the bid on was actually vauluable at assisting conversions for his other other keywords. That’s lost opportunity.

You don’t know what you don’t measure.

Everything looks good if you look at it through the lens of the “right metric”. I think the best policy when managing advertising campaigns is to first remove any rose tinted glasses you might be wearing.

Fact is that conversion is an important metric. Each every channel in an advertiser’s arsenal should be evaluated by how it converts individually as well (as Peter pointed out) as how it contributes to the complete marketing mix.

It is correct to say that often advertiser’s fail to use CPM advertising for what it is designed to do (increase branding, develop awareness, create uplift in other channels), but it is misleading to say that display advertising “performs well” by itself.

Advertisers who have the technical capability and the marketing know how to maintain a mix of display advertising in their marketing arsenal should do so even in a down economy. Many don’t. So it is hardly surprising that in a recession they turn away from a channel that simply burns money when improperly used.

Angel

By the way Peter, great post.

So when various groups predict a down turn in display advertising because advertisers are looking for better performance based advertising, I’m shocked. The reason why I’m shocked is because display advertising actually does perform very well. Advertisers just aren’t using the right metrics to understand the full performance of their display ads. While display ads might not always convert well on their own, they do perform very well at driving conversions, traffic, and brand awareness to other channels of online advertising such as search. By not running display ads, advertisers could actually be risking a 22% + lift in conversions.

For more information about this, check out my post on the Official Yahoo! Search Marketing Blog at: http://www.ysmblog.com/blog/2008/11/03/measuring-search-and-display-for-success/