• by Seana Mulcahy , May 17, 2004

One of the hardest questions I get asked is, "How do you know when we've saturated our target audience?" Coupled with that, it seems as if there is always a surplus of impressions out there to be had. Media folks (like me) get wide-eyed when our client's spends are increased. Is it a challenge to "own" and audience? Or, is the challenge obtaining an accurate saturation reading from the dusty old crystal ball?

Years ago, someone did a study (I forget who now) that mapped out "banner burnout." For those of you who are not familiar with the term, I'd loosely define it as the point where an audience becomes saturated by a given piece of (exposed) creative." For branding purposes, we used to use the rule of thumb that you want a prospect to see your creative at least 3 times to resonate with them.

I've been waiting for a solid study to come out for a while now. Several months ago, Young-Bean Song, Director of Analytics of Atlas DMT, spoke at a Boston Interactive Media Association event that I chaired. He teased us by saying Atlas would release such a study this spring.

Young-Bean has been speaking about its findings. He says that on average 33% of all online ad impressions occur after a user has already seen a campaign 10 times. This results in a waste of precious ad dollars.

Yes, I agree it is still confusing. Let's think about this for a moment. My brain automatically wants to perfect impression count. However, what Atlas has found is that impression distribution for most campaigns tend to be completely out of whack. For instance, a small group of an audience could be completely overexposed whereas the bulk of your target audience may not be seeing your ads enough.

This is a red flag to make sure your data analytics team is carefully assessing impression counts. This is where the term "optimal frequency" comes into play. Optimal frequency refers to the ideal impression level for a given campaign. Simply put, you don't want to overexpose an audience and waste money or underexpose (and waste money). Certainly, there are separate answers for those ads of branding versus direct response. The funny thing is, most mentions or studies have been related to branding.

As far as I know, the Atlas study is the first to address direct response. Atlas took a look at 38 companies that ran campaigns Q2 '03. They then took a look at click-based conversion rates at each frequency level. It was determined that the first 3 impressions for each advertiser had a 100% lift on average. More than half the advertisers experienced their highest conversion rate on the first impression.

Now, I don't want you to take this out of context and think you only need to show an ad to each prospect one time only. There are a lot of other factors that no one could precisely track. For instance, what's going on in offline media when your campaign is running? Are your prospects seeing the same or similar messaging elsewhere? If so, could this be giving your campaign a lift?

The bottom line is, Atlas says it right when they point out that at any given time there are only a fraction of users who will immediately respond to your ads. Check out the study and let me know what you think. Post to the Spin board and no, Atlas is not a client.