Where You Should Stick Your Ad and Why

Return to Tarred by the broad brush

I'll start this section by stating that I'm sure there are existing ad placements that do very well. Let me also state that I was very surprised to find out how the media buying and planning process worked. I found out in conversations with Avenue A| Razorfish, Casale Media, FindMeFaster, ICON International, Underscore Marketing and others that agencies aren't told much about the material they're placing on a web page.

More accurately, they're not told things that I thought were obvious considerations: what does the ad look like? What else is on the page?

In some cases they're provided a product or product category and that's about all.

Let me ask if the following approach makes sense to you: I'm going to tell you what you're aiming at but I'm not going to tell you if you're using an arrow, a spear, a pistol, a rifle, a tank, a knife -- whatever -- to hit it. And I'm also not going to tell you what's between you and the target, hence what's going to get in the way of aiming and hitting the target.

However, I still want you hit the target with a very high degree of accuracy or I'll find someone else to take potshots for me.

Ad placement ROI could be astronomically better if media buyers and planners could answer a different set of questions. Here are some questions that might make no sense to agencies yet are revealing to cognitive psychologists and cultural anthropologists. The placement suggestions which follow them are from NextStage’s and related research:

  • Is the target audience active socially?
  • Do they have a strong social network?
  • Do they have a large (extended) social network?

Believe it or not, any of those questions when tied to gender strongly suggest the use of marketing to women with either a 300x250 Medium Rectangle or a 160x600 Wide Skyscraper, to men with either a 728x90 Leaderboard or a 120x600 Skyscraper.

  • Is the ad going on a personals site?
  • Is the target audience in a relationship with someone?

Questions like these are relevant because recent human functional neuroimaging studies of romantic attachment show that affection-based relationships are tied to reward seeking behavior.

This translates to marketing to females via either a 120x600 Skyscraper or a 300x250 Medium Rectangle, to men with either a 160x600 Wide Skyscraper or a 300x250 Medium Rectangle.

Thus both genders will respond more favorably to an ad on a personals site if the ad is a 300x250 Medium Rectangle than any other form.

  • How many times will the target audience experience this material in any form in a given time period?

This answers which placements should appear when during a given campaign and when responses to the campaign will begin (NextStage is currently working on a method to determine how many times an audience needs to encounter something in order to respond favorably to it).

  • Is there a social (viral) aspect to this campaign and if so, when will it be enacted?

This answers how often a given ad needs to appear in the target's visual field in order for the campaign's message to take hold.

  • Is there a blog or wiki element to this campaign?

Similar to the above, this answers how often a given ad needs to appear in the target's visual field and the order in the visual field hierarchy in which the material needs to appear.

  • If this is a multiphase campaign, how long after phase 1 does phase 2 begin?

This answers exactly what needs to appear in the ad and where in the visual field the ad needs to appear. Information (an ad) that repeatedly appears in a single position over time causes the meaning of the information to lose impact to those seeing it, what is called "wear out" in the trade.

Knowing when different elements in the same campaign go live strongly dictates where ads need to appear in order to increase recognition and reference in the target audience's mind.

  • What are the target audience's values (religious, ethnic, cultural, historical…)?

This answers where in the ad the branding element needs to appear because the ad itself is not the brand, it is a vector for conveying the brand and creating action associated with that brand.

My awareness of the gap between what NextStage's researchers would think to ask and what media buyers and planners could answer grew out of conversations at IMedia's Dec 06 Agency Summit in Scottsdale, Arizona, and documented in my blog post, Follow Up to "On the road again..."

As FindMeFaster CEO Matt Van Wagner said, "I definitely have a better idea of the audience than I ever had before."

Part of the fun of this research and developing the tool that grew out of it was figuring out what agencies knew that we could translate into data for the necessary calculations. But it's not an easy process.

Next: Let's stop talking about "behavioral"

Joseph Carrabis is CRO and founder of NextStage Evolution and NextStage Global and founder of KnowledgeNH and NH Business Development Network. Read full bio. He was recently selected as a senior research fellow and board advisor for the Society for New Communications Research.

 

Comments

Joseph Carrabis
Joseph Carrabis August 28, 2008 at 8:30 AM

Rob, Sorry I didn't see this until now. "Many actionable solutions"? Well, we do have that tool that determines where ads should be placed based on user responses to some questions.
Are there some general rules? How general would you like them? "For boomer males do x except when marketing high end cars or when there's a slightly younger female in the creative or the product is an OTC or ..."
I know the goal is to make things as simple as possible and sometimes the simplest solution results in the 30 or so questions we currently ask (and I'm happy to learn there's a simpler solution with as high an accuracy). This isn't a defense (as I hope you'll agree I'm always working to improve things and enjoy discovering alternate methods that simplify things and freely recommend them).
Developing the tool and this article was quite an experience for me. I was shocked to learn about the disconnects that exist(ed) in the system -- who didn't talk to whom, who didn't know what and couldn't find out. Something as simple as not knowing where (on what sites) the creative would show up, hence essentially shooting in the dark and hoping you were at least shooting in the direction of the target. You reference this in your "hysterical/horrifying ad/story combinations".
"...there are still a number of moving parts that need to be addressed..." Agreed. And quite willing to discuss.
And thanks for reading my column. - Joseph

Rob Graham
Rob Graham June 22, 2007 at 1:39 PM

I enjoyed the article and found is very interesting. I'm a bit biased perhaps as Joseph is a friend of mine. However, from a practicality standpoint I'm not sure he was able to offer many actionable solutions. Joseph has accurately pointed out that there are a number of shortcomings regarding ad placement that few of us would dispute. However, I don't think he is able to, not would I expect it, to offer up some sort of unifying solution here. I think the real challenge is to get advertisers to think about the benefits of really planning ad buys and ad placement and not just defaulting to choosing a group of silos to push content into. Also keep in mind that most of the logic engines behind contextual advertising still haven't been tweaked enough to avoid the sometime hysterical/horrifying ad/story combinations that result. The article wasn't too long and it brought up some excellent points for media planning. However, there are still a number of moving parts that need to be addressed further before the industry can really act upon a lot of these insights.