My last post that was in response to Tom Hespos' published article in iMediaConnection, "How to Pick an Ad Server," the post generated a lot of interest.  People don’t seem to leave comments on my blog but they do email me, which I always appreciate.  The feedbag that I have received is that people want to know more about the experience I have with buyers, about selecting ad servers and the implementation process one experiences after a selection has been made.  That last step I can comment on based on clients I work with directly both with TruEffect and the feedback I have received based their previous experiences with other ad servers.

The agencies and advertisers (mostly agencies) that really put the thought into selecting an ad server shop the industry.  Usually they start with the Big-3: DoubleClick, Atlas and Mediaplex, and then have one or two others like TruEffect, [X+1] (formerly Poindexter) or Zedo added to the list.  Poindexter is not actively getting the consideration it used to get ever since shifting its business model to web site conversion marketing.  Their ad serving technology was never as competitive with features (especially reporting) in comparison to the big three, but they do have some major accounts. 

When shopping the industry, I have seen agencies use RFPs, or more informally collect proposals in order to build a matrix of features, benefits and pricing.  The most commonly evaluated comparative points are Interface evaluation (subjective), Rich Media capabilities, Targeting capabilities (geo, day-part, cookie), Behavioral Targeting capabilities (event-based of course), Optimization capabilities, Reporting capabilities, Price, Training, Customer Support and Implementation.  I am going to break these sections up into separate posts.

The Interface Evaluation takes into consideration the level of complexity of the design.  DoubleClick and Atlas are notorious for lacking intuitiveness in their navigation.  Both tool sets offer pretty much every capability that an ad server has to offer and they stick it all in there for the user.  The organization has been band-aided over time so as features have been added the pages have lengthened and more buttons and fields have been added, which makes it harder and longer to setup a campaign.  The biggest complaint that I hear from people when evaluating these tools is, “wow” that’s going to take a long time to learn.  What I hear consistently from established DC and Atlas account users is that training requirements are high.  Each new FTE needs to be adequately trained and that training always has an incremental cost.  Interface complexity is important when you consider the speed by which you can setup and get a campaign running.  Let’s face it, by the time you are ready to get a campaign live it’s already the 11th hour and time is of the essence.  Both DC and Atlas are single-entry systems which means each insertion must be keyed-in by hand, each creative must be matched to each insertion manually and then the schedule must be activated in order.  The bulk-upload features are clumsy and not widely used.  Efficiency is not inherent in the online applications.  Don’t get me wrong, as the commonly used platform in the industry, most people are very proficient on DC’s platform but there is a learning curve.  Having established experience using DART is a good resume skill!