Alex Baxter Frank Cutitta Paul DeBraccio Judy Franks Dave Hendricks Marko Hurst Jay Lauf Daniel Lagani Karen Macumber Diane Salvatore Ken Sonenclar Domenic Venuto Marta Wohrle Matthew Yorke |
|
Paul DeBraccio
In the 15 years I have been in online advertising, the term optimization means one thing to me. It is when an ad agency analyzes the sites they are using and culls the weak ones while bringing in new ones that have potential.
That word has come to mean terror or glee to digital publishers as it can cost them money if the optimization goes against them. A little known fact is that optimization costs the agencies money, too, and that cost is increasing more and more these days, and most media directors dread this necessary and tedious task.
Most of the more sophisticated digital agencies have a few processes that need to be completed to optimize their ad buys for clients, and most of them have to be conducted manually. When a client asks, “How much was spent and where was it spent?” most agencies cringe and move forward to aggregate the data—manually. These systems needed to implement and manage digital campaigns are complex but necessary. The list begins with the ad server and audience measurement programs. Not to be left out are ad verification services, other targeting programs, ad exchanges and systems like Atlas and Mediavisor. These are all useful and well-designed systems, except that they do not speak to each other and they all provide different numbers. Often they are 30-40% variances. These systems also want to own the metadata too. The disaster is that this is a process that demands minimum errors as there are millions of dollars in play.
That means that staffs at agencies are charged with manually reading this data and making major decisions about a campaign by combining data and, for the most part, extrapolating. As you can see, this means a lot of late nights for those assigned to these tasks and many costs assumed by the agencies.
If an agency is implementing a campaign internationally, the aforementioned problems increase exponentially. A campaign scheduled for the U.S., Europe and China presents problems that agencies must attempt to fix—you guessed it—manually. It does not seem like there are any technical fixes in sight either.
Vivaki is good at aggregating many data points and may have a chance of leading the way to streamline this process with its overall Pool strategy, but when that will happen is not clear at this time.
This is not a trivial issue to major online media directors either. In traditional media departments, the reconciliation issues require staff to run actual ratings numbers versus estimated ratings for TV and radio and then approve the billing or issue a revision. (At least that is the way it used to be.) In fact, a digital media director strongly suggested that I write a blog on this topic as it is becoming more and more complex and expensive. Based on her suggestion, I queried a few others for their opinions too. They agreed that this may be one of the more difficult aspects of their jobs.
Publishers are not too happy about this process either. When we describe it to new publishers in my consulting business, they look at us incredulously. They feel that when they receive an IO from an advertiser for a six-week campaign that it should run until the end, no matter what. Of course, optimization is a logical and proper strategy for both advertisers and publishers, but the current infrastructure is at the breaking point financially and practically.
Perhaps the IAB or some other online organization can form a committee to address these issues. The IAB has done a great job with the issues of ad sizes, T&Cs, video and other aspects of the business.
I, for one, would be open to help find a solution, as we will all benefit. The celebration will be led by those assigned to compile this data.
Minsider Paul DeBraccio is CEO of Interevco and former head of sales for GeoCities and Tripod. Paul has also been a media planner and magazine executive at Omni, Hachette, Wenner Media and Condé Nast.
|
|