AD NETWORKS
Published: May 18, 2006
Q&A with ValueClick's David Yovanno (Page 2 of 5)
 

Ad networks and branding; plus, is there an inventory crisis?

Berens: This is fascinating. Let's dig down into the brand marketing/direct marketing distinction on the ad network side. People think of ad networks as a great place for direct response. Lately, we have been hearing about lots of different channels that people will not initially think of as appropriate branding channels. Search, for example, as a branding platform was something we heard about. I do not know if you saw Ron Belanger's presentation at the last Brand Summit, where I know we both were…

Yovanno: Yes I did.

Berens: Let's talk a little bit about what ad networks can do for brands.

Yovanno: Sure. We just published an article related to this specific topic, and it is based on the presentation that I presented at the iMedia Brand Summit in January. Again this discussion needs to be framed based on the objectives of the advertiser. So, when we talk about "brand related objectives," what we are typically talking about is driving lift in metrics such as: awareness, purchase intent, message association and brand favorability-- all of those things that are measurable through the ad effectiveness research studies that Dynamic Logic or Insight Express offer.

That may be a primary objective for a lot of our brand-marketing clients, but there is still almost always a secondary objective for leads and sales. I do not think I have ever worked with a client that is just only measuring those brand metrics.

Berens: Why shouldn't there be?

Yovanno: Exactly, but, for those that have a primary objective to drive lift in these metrics, it is pretty simple. In fact, we employ a lot of our direct marketing techniques, the skills that we developed with our direct marketing clients over the years. We apply these same principles to our brand marketing clients. For example, if we determine up front that the key metric that we are driving is intent to purchase, then we set the campaign up in a way across our network to measure that. We are able to see what creative and what placements within our network drove the highest response within that brand metric. And, we are able to optimize the campaign in that direction.

So, to answer your question more specifically on the point of critical optimization, which is a pretty popular buzzword in direct marketing, and definitely applies to brand metrics, you find ad networks that have the critical mass, the technology and the skill to manage ROI in that way… it is a great vehicle for that.

The other thing that I point at is just the complementary reach that is available with ad networks. The most common media plan that I promote to clients is for them to work with a major portal like Yahoo, work with a top vertical site that matches your audience like ESPN or iVillage, and then leave the rest of the internet to an ad network like ValueClick Media. We specifically are unique in that none of our inventory comes from any of the major portals or top vertical sites today. So, we manage in a sense what the essence of the internet is really about, and that is fragmented sources of content and reach. We have a tremendous group of quality sites within the network. And, we are able to offer unduplicated reach to tens of millions of users that can't efficiently be reached anywhere else. That is a strong argument for using a true ad network. 

Another big one for brand clients is just the limitation that exists within things like rich media executions, because of the heavy frequency capping. For instance, you cannot do a large, rich media campaign on the front page of Yahoo. They are very sensitive to turning away their users, especially on the home page, so there is a frequency cap of one, let's say, in the case of a take-over ad on the home page of a site. 

There is a lot of buzz around behavioral targeting; and, the results are there. The challenge seems to be having enough data available to actually track enough behavior for it to be meaningful and then have a marketplace available to you to efficiently retarget users. So, when you talk about knowing more about the user, and who you are reaching with your message, if you are working with a single site on a behavioral program there is just not a whole lot of scale available.

When you consider the marketplace of the data that a network can pull together -- like browsing behavior, search behavior, shopping and purchase behavior and other ad-response behavior -- there is just a greater opportunity for data when you consider the number of sites, impressions served, unique users, et cetera. A network is pretty much the only model where you will find any kind of scalability with a behavioral execution.

Furthermore, targeting aspects like, "time of day," or "geography," those things have a lot of limitations in terms of how much inventory is available from any one site, or even on a portal. And again, because an ad network typically pulls together more sites and more users, you are going to find more scalability for those targeting options.

Is There an Inventory Crisis?

Berens: You have mentioned "inventory" in several interesting ways in the last few minutes. One of the questions that we have been asking, and getting a lot of different perspectives from a lot of different people on, is whether or not there is an inventory crisis. Lots of people, particularly automotive marketers, think there is. I would suspect that you would not think that there is, but maybe I am wrong.

Yovanno: I agree there is a crunch with what is considered "good" inventory. There always has been. That is why true ad network models providing valuable information on what is known about the user opens up more inventory. There is a lot of run-of-network inventory that exists on the internet and there are a lot of impressions that are going unused. But, we are in a unique position in terms of the data that we have available, and the way that we compile it to know something about the non-personally identifiable behaviors of the user, making more of our inventory more valuable because we know something about the user's behavior that other companies do not. And we haven't even begun talking about what data ValueClick, as an enterprise, has available.

Berens: We are getting there.

Yovanno: So, we can save that for later. The bottom line is-- there is a ton of inventory that is not being used on the internet because it really is being treated as run-of-network. Not a whole lot of people are in the market for the ads that are showing up on free email accounts, for example.

Next: The ValueClick merger with Fastclick

White Paper Library

View More Research »