The Next Evolution Of Programmatic: The Publisher Exchange

The Sell Sider” is a column written by the sell side of the digital media community.

Today’s column is by Paul Bannister, co-founder and executive vice president at CafeMedia.

2019 promises to be the beginning of the next transformation in digital ad buying.

RTB-driven programmatic advertising started in the late 2000s (2007-2009) and drove the initial burst in data-driven buying, particularly with the release of the RTB 1.0 standard. Header bidding started gaining popularity in 2014 and changed the way buyers accessed inventory, how publishers prioritized their ad stacks and shifted programmatic from being a vehicle for low-quality inventory to a truly premium channel for buying digital media.

Now another technical change is on the horizon that promises to remake programmatic again: publisher-direct connections or publisher exchanges. Until now, almost all programmatic buying has gone through a similar pipeline:

Advertiser -> demand-side platform (DSP) -> exchange -> publisher

This setup has served the industry well, and its design has led to its explosive growth. But it may be reaching the end of its usefulness, at least in some scenarios.

The basic idea behind publisher-exchange connections is for DSPs to connect directly to publishers, bypassing ad exchanges entirely. This change reduces the number of “hops” between buyer and seller, trims fees, lowers DSP costs via reduced queries per second, increases transparency and improves cookie matches. In theory, all of this will lead to better performance for buyers and higher revenues for publishers, although this needs to be proven out by the early adopters.

What is basic at a high-level is much more complex in the details. Exchanges perform critical functions such as inventory packaging, auction optimization, ad quality, reporting and more.  These functions must be replaced by either the DSP or publisher, which puts a significant amount of technical burden on those companies beyond their existing businesses, and some may not be prepared for this commitment.

Exchanges won’t stand still and watch their market erode. Some will increase the services they offer, providing higher value to buyers and sellers. OpenX, for example, has been pushing toward this with its acquisitions. New low-cost entrants, such as TripleLift’s Header Direct product, will push other players to reduce their costs. And AppNexus and Google, via their owned DSPs, and others will build out stronger unique or prioritized demand that will preserve their positions of access.

It’s also unlikely that the endgame of this transition is the elimination of the exchange. Rather, it will accelerate the demise of marginal players and consolidate power with a few strong players that will remain partnered with all publishers, including those that have moved heavily to the new publisher-direct model. A typical publisher in the next few years might have two to three exchange connections and 10 to 15 DSP connections.

This potential disintermediation of exchanges points to where the next true battleground of supply-side programmatic will be: the wrapper. The place where the header-bidding auction occurs, the wrapper is likely the platform through which publishers and DSPs will interface with each other. Index Exchange foresaw this development years ago, and it’s a direction that Rubicon has also recently considered. There are a number of other companies in the space with their own wrappers or ad-management solutions also positioning themselves for this platform change.

Over time, these direct connections will also reduce access for marginal DSPs. As exchanges lose access (and priority), DSPs that can’t connect directly to publishers will lose access, forcing buyers to move to the more connected platforms. This new publisher-exchange world will force the industry consolidation that has been predicted for many years.

Publisher-direct may also be the only route through which the most premium inventory flows.  Large-format video, particularly full-episode player video, is the most sought-after type of ad inventory and has mostly been off-limits to the ad tech ecosystem. Major media companies have mostly sold it directly to buyers, but as the inventory pools grow and buyers demand programmatic access, it has become the most important pool of inventory for exchanges to bring online.

Since this growth in inventory will occur simultaneously with the transition to publisher exchanges, it’s possible that many premium players will decide to simply bypass the exchanges entirely and go directly to the DSPs. This maximizes media companies’ revenues and gives them complete control over access, which is a major concern of releasing inventory onto the open market.

In 2019, DSPs and large publishers will begin this transition. Like header bidding and the birth of programmatic in general, the first year or two will only see a few publishers and DSPs making the jump to this new approach. The investment required and complexity of this new approach will require a “go-slow” approach to ensure that things are done correctly.

But when we look back in a few years, 2019 will clearly join 2007 and 2014 as one of the years when a major tectonic shift in programmatic took place.

Follow Paul Bannister (@pbannist), CafeMedia (@CafeMedia_) and AdExchanger (@adexchanger) on Twitter.

1 Comment

  1. R McGrory

    The end state here is ad/data tech facilitation of an insertion order b/w a buyer and seller where the all in take rate of 2-4% (slightly more if it include sales representation). Anyone building a business assuming more will be disrupted. Happy 2019!

    Reply

Add a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>