Explaining the Header Bidding Juggernaut


IT’S HARDLY POSSIBLE TO HAVE an advertising industry discussion in 2016 without talking about header bidding.

The merits of the header bidding solution have been discussed widely from trade publications to conference stages, and seemingly everywhere in between, but usually the coverage focuses on the simplest outcomes.

“This will change the game for programmatic, allowing buyers and sellers to truly transact on valuable audiences more seamlessly.”

In light of all the opportunities the solution creates for publishers, header bidding is certainly a technology worthy of attention. When implemented on publisher websites and apps, header bidding can increase fill rates and raise the average revenue publishers see from auction impressions. But that’s not where the largest opportunity lies.

The real sea-change we’re about to discover is this: by exposing every impression to programmatic demand, header bidding will enable publishers to have better control over how they monetize their entire inventory mix, and over what types of opportunities they can create for buyers. In particular, it will allow publishers to control prioritization of private marketplaces and automated direct deals with buyers. This will change the game for programmatic, allowing buyers and sellers to truly transact on valuable audiences more seamlessly.

This is an important point to under score. Header bidding is a technical means to a business end: better monetizing all inventory. So when evaluating the merits and success of header bidding, let’s keep in mind that what we really want to achieve for publishers is higher yield, and for buyers the ability to truly buy under a model that allows them to leverage data, buy audiences at scale, and on a guaranteed and/or forward basis when desired.


The solution is in such high demand because traditional ad servers can’t ofer publishers the controls they need to take advantage of today’s digital marketing opportunities not without fundamental changes to how ad servers work.

At almost 20 years old, ad serving technology, which does publishers’ bidding in terms of how it prioritizes impression opportunities for a multitude of eligible buyers, didn’t anticipate real-time bidding, automated guarantees, and other novel ways marketers can now buy digital ads.

As a result, ad servers constrain publishers’ ability to take full advantage of today’s advertising technologies. They also force publishers to make unnecessary trade-offs in terms of volume and price when they do use those technologies.

Two vestiges of the traditional ad server paradigm are particularly illsuited for the era of ad automation.

First, since indirect deals those not negotiated directly between publishers and brands were once low-priced remnant buys by definition, it made sense in years past to relegate them to the lowest rungs of ad server prioritization. The objective of those indirect deals was to keep publishers from leaving money on the table by letting them monetize ad space they simply couldn’t sell directly to brands at market rates.

But those “indirect” ad buys have evolved into “technology enabled” buys campaigns that can represent the full spectrum of revenue opportunity and customization because buyers are utilizing data and publishers are willing to expose more supply knowing that an increasing level of priority and calculation is being brought to the buy.

Secondly, the low priority, indirect deals of yesteryear were sold at a fixed price, at a time when it was ad networks that mostly filled publishers’ leftover inventory. Now that buyers are able to bid for those impressions in real time, based on the predicted value of the consumer who views them, some of those indirect buyers submit prices approaching top-tier rates.

Unfortunately, the fixed price para – digm locks those potentially high-dollar buyers at a lower priority level in the ad server, artificially blocking opportunities for them to compete with high-priced, direct-sold campaigns.

Both of these traits of traditional ad servers suppressed publishers’ ability to increase their revenue by maximizing high-value and technology-enabled demand. Traditional ad servers also deprive willing buyers access to high-value consumers on those publishers’ sites and apps even when they were willing to pay top dollar for the privilege of showing them an ad.


Then along came header bidding (a much older solution than a lot of people seem to realize) to make ad servers work for the current time. Header bidding is akin to a plugin or adapter for the ad server: it lets publishers that use traditional ad servers get around the artificial constraints on how they structure their inventory and prioritize buyers’ access to it.

In the simplest terms, the mechanism that makes this possible is the “pre-auction.” The page or app receives pricing information for all eligible buyers as one of the very first actions that occurs when a page or app loads. This makes all the difference for how the ad server can make its decisions.

The results? Programmatic demand can now compete with direct-sold campaigns at any price or priority within the ad server. Private marketplace deals (deals that are pre-determined between buyer and seller but in which the individual impressions are bought in the auction) can now sit higher in the ad stack and enjoy a higher rate of audience matching. Real impression values can now be used to determine prioritization of RTB campaigns instead of estimates and averages, fostering fair competition for every impression, and giving high-bidding buyers a real chance to compete.

And publishers can now control how all of these pieces interact with each other.


Given both the excitement about header bidding and the complexity of the implementation, it’s not surprising that some of its advantages have been misstated or oversold.

”Where header bidding will make a real diference for publisher revenue and relationships is in its ability to prioritize automated guarantees and private marketplace deals.”

It’s fairly common to read, for example, that a primary advantage of header bidding is that it makes programmatic supply sources compete for impressions, thus driving up revenue for publishers. But what should in fact happen over time is we reach an equilibrium where the right demand meets the right supply at the right price. So high-value audience impressions will garner more, and the lower will see the opposite trend.

And while a thoughtful header bidding implementation and ad server setup can increase opportunities for high-bidding RTB buyers to win impression opportunities, I think too much emphasis is placed on this outcome, obscuring a decidedly more consequential advantage of header bidding: where header bidding will make a real difference for publisher revenue and relationships is in its ability to prioritize automated guarantees and private marketplace deals.

As more publishers awaken to that advantage of header bidding, I think our industry will more clearly see where the true promise of the solution lies. After all, header bidding is about giving publishers more control and allowing demand to flourish without technical limitation.

Ultimately, header bidding is about optimizing access to the marketplaces that offer both scale and high quality programmatic demand. Those marketplaces that provide both buyers and sellers the best-in-class tools to manage the full spectrum of deals, have the most high-quality publisher inventory, and have the audiences at scale across devices will win.

Download your copy of The Programmatic Mind Issue #11 at