The US' Interactive Advertising Bureau (IAB) and Bain & Company recently released an interesting study showing that "the use of 'ad networks' surged from 5% of total ad impressions sold in 2006 to 30% in 2007."
This isn't good for online publishers, according to Bain & Company's John Frelinghuysen, the author of the study:
"Online publishers are producing more inventory than the market demands, and risk devaluing the premium nature of their brands, particularly in light of ad networks growth and their dramatically lower pricing."
According to the study, branded publishers have the most to lose:
"As online publishers continue to experience growth rates of 20-30% in ad revenue, the race to create new advertising opportunities has left publishers with an excess of inventory which they are selling through ad networks at up to 90% discounts versus direct sales rates.
"The study finds the trend particularly foreboding for branded online publishers who traditionally earn between $10-20 CPM (the industry term for the cost per 1,000 ad impressions advertisers pay) and therefore risk severe price erosion."
The study recommends, amongst other things that:
- "Publishers must become more disciplined in managing ad inventory and deploy improved methods and tools to enhance yield management."
- "Ad networks should partner more closely with publishers to enhance the value of the relationship for both parties."
While the conclusions are not surprising given the involvement of the IAB, I do think they make sense.
In a past discussion on ad networks, it was noted that some branded publishers, such as ESPN, have already made the decision not to use ad networks, and others are reportedly thinking of following suit.
This is not entirely surprising given the often substantial difference between the prices received for ads sold directly and those sold via ad networks.
In my opinion, some of the problems highlighted by the IAB study result from a natural conflict between the interests of branded publishers and ad networks.
Ad networks, with their focus on scale and automation, have a model that is the antithesis of the direct selling model that most branded publishers receive the greatest benefit from.
The direct selling model, however, does have its limitations and for many publishers, there is a place for ad networks.
Yet the interests of the ad networks are in many ways more closely aligned with media buyers than they are with the publishers they represent, and when media buyers can purchase ads on branded properties for a fraction of the "list price" through ad networks, branded publishers are the ones who get the short end of the stick.
So what's the solution?
Unfortunately, given my experience with large ad networks, I think ESPN's decision to abandon them altogether may make sense for a lot of branded publishers.
In some cases, the benefits of selling remnant inventory through an ad network do not outweigh the benefits of holding firm on the value of your properties and focusing your energies on maximizing direct sales.
Yet I also agree with Frelinghuysen and believe that there is opportunity for ad networks to "partner more closely" with their publishers.
In my experience, ad networks, especially the larger ones, treat publishers like commodities, not clients.
Some publishers are commodities but branded publishers aren't and I think it's time for ad networks to realize that.
While I'm hesitant to encourage the creation of more ad networks given that I believe there are already far too many, I think there might be some room in the market for niche, specialized networks catering to higher-caliber publishers and which act more as an extension of those publishers' in-house ad sales teams than they do as an ad network.
Such a model could conceivably more closely align the interests of the ad network with its publishers, enabling branded publishers to maximize their inventory without minimizing the value of their inventory.
There are some companies out there pitching such a model and I'm preparing to enter into an agreement with on,e so I'll be sure to revisit this topic in the future as warranted.