Michael Zimbalist is vice president of research and development operations at The New York Times Company (NYSE: NYT). Previously, he was president of the Online Publishers Association.

A fissure opened within the online advertising community last month following the release of new research by the Online Publishers Association. Drawing on three years of Dynamic Logic data, the OPA report concluded that ad campaigns on premium content sites produce greater lifts in favorability and purchase intent than campaigns on ad networks. The Wall Street Journal characterized the reaction to the report as industry infighting amid a shrinking pool of ad dollars. Others were less generous, accusing the OPA of a conducting a scorched-earth policy, saying the OPA had turned its back on legions of “long tail” publishers, many of whom arguably achieve the same level of editorial quality as other, more prominent OPA members. 

So let’s unpack the infighting. At its root is the view that the OPA, as a proxy for premium publishers, is waging a war on ad networks and is failing to acknowledge that networks and publishers play different roles in the online marketing mix. After all, many networks are supporters of the OPA because they supply a service to OPA members: It is well known that many “premium” content sites clear “non-premium” inventory through networks.   

What’s left unstated by both sides in the debate is that premium websites, whether OPA members or not, generate both premium and non-premium ad inventory. There is elasticity to the supply of online advertising impressions that makes this inevitable. For example, when Michael Jackson died, the flood of visitors to news sites generated more ad inventory than any direct sales force could liquidate at premium rates. Network deals for occasions like this can be beneficial. 

Branded publishers often overlook the existence of non-premium inventory within premium sites. What’s worse, however, is the failure of networks to acknowledge the distinction between premium and non-premium ad positions within the branded publisher sites burnishing their rosters. The fact is, while networks may have access to certain non-premium inventory, they can never sell what is rightly called premium. (Premium inventory is fully guaranteed. The terms of sale, including placement and flight dates, are completely transparent, and it is sold direct, through negotiated sales.) Marketers recognize this. So do media buyers. By the way—it’s the job of the OPA to spotlight this distinction, just like it’s the job of CAB to promote advertising on cable as opposed to broadcast, even though it’s all just TV!

As their supply of impressions has increased, many large publishers have gravitated toward the network option to generate a small but reliable revenue stream alongside direct sales. The key questions for publishers working with networks are: Will the channel conflict cannibalize direct sales? Will it erode my pricing power? These questions elude easy analysis. Many publishers are also asking themselves: Should we withhold inventory from networks? Can we eliminate some ad positions entirely to create more scarcity? Can we innovate to build new ad products that are not easily commoditized?

The current pace of innovation in online display advertising is unprecedented, and the bulk of it is focused on the non-premium space:  Demand side networks, ad exchanges, data exchanges, and the emerging infrastructure to accommodate real-time bidding are pushing us to the brink of an entirely new paradigm.

The opportunity for publishers is that real-time biddable exchanges will help make non-premium inventory more like premium by matching unique parcels to buyers who have been heretofore unreachable via direct sales and for whom the inventory is worth more than the remnant rate. Not only will this increase publishers’ revenues—it will simplify the marketplace for advertisers. Premium positions will be available via negotiated direct sales, while spot buys will be transacted within exchanges in real time. Publishers need not fear such an outcome, because even in a real-time transaction scenario, premium display will remain a valuable tool in the marketers’ online arsenal, one whose meaning is clearly defined and whose impact is well-documented and understood.