Last week, our editor Masha Geller talked about the commoditization of banner serving. I'd like to take this one step further. The "banner serving industry" is headed for a fall if they do not effectively redefine and communicate their business model. Let's look at what we get out of banner serving and what is important. There are 8 important functions involved here, including:
Analytics Optimize, based on analytics
In addition, there are a number of considerations when comparing the companies performing ad serving, including:
Bandwidth/Access Speed (Latency)
User interface (UI)
Data mining capabilities
Please note that the "serving" of banners or trafficking is just one function of an ad server. In fact, it could be argued that it is the least important. If you look closely at the future through your crystal ball, one has to ask why we base all of our metrics on either a request to serve a banner or the sending of a banner to a Web page. Why don't we measure the consumer OTS (opportunity to see) on the client desktop or browser? This would make a Web advertising impression equivalent to that of other media, in line with the classic ARF communication model.
In fact, a company called Solbright does just that. And they don't even serve banners! The banners are served by the individual sites (just like the old days) and Solbright, through the patent they license from Thinking Media, measures what the consumer sees on their computer. Whether the message comes from an "ad server", a cache, or is hard coded onto the page. The logic and simplicity of this is overwhelming.
So what is important? First and foremost, are metrics and the analytics that produce the analysis that assists in our decision path as to what to do next. Which sites to use, which creatives units and messages to deploy, which targets (either demographic or sociographic) are responding, etc. And, it is NOT about click-through. It is what happens after the impression is registered, whether or not the consumer clicks on the ad. (Important to note that, according to research from Engage/AdKnowledge, DoubleClick and other sources, some 35-40% of all consumer ad driven interactions are post impression, not post click.)
Just as we have walked away from hits, we should ban the term clicks or clickthroughs as irrelevant. (Have I just eliminated a category of sites, CPC networks?) With reliable post-impression data and post-click data, maybe we should not count clicks at all. Issues like the time lapsed between impression and activity on the site are much more relevant.
Within metrics, optimization (both of sites and creatives) is the single greatest point of leverage that a media person has in the Interactive buy. More so than planning or buying. Optimization can increase the efficacy of a campaign by 3-5 times in