AppNexus president Michael Rubenstein on the header bidding shakeup

AppNexus president Michael Rubenstein on the header bidding shakeup .

The debate over the merits, and shortcomings of exchange-based media trading continue, and the emergence of header bidding – and Google’s bid to counter it - is still one of the defining questions for publishers to face in 2016.

AppNexus has been at the forefront of this header bidding debate, eager to point out the supposed bias in earlier ‘waterfalling solutions’ to Google’s DoubleClick.

Although it must be said that a raft of others including AOL, OpenX, Criteo, etc. have also helped lead the charge, and the recent $200m mark down of Rubicon Project’s stock price for its late entry into the game demonstrates its importance.

Earlier this month, AppNexus doubled down on its header bidding

strategy with the launch of PreBid Enterprise.

The launch of this open source ‘wrapper’ technology is integrated with multiple sources of demand, and ultimately, to ‘avoid the Google AdX tax’.

The Drum put further questions to AppNexus president Michael Rubenstein.

PreBid Enterprise:

The Drum: Is this need to further develop the technology coming from publishers? And if so, is AppNexus looking to roll it out live with any publishers for instance?

Michael Rubenstein: The demand for header bidding technology and services is coming from publishers. We currently have approximately 225 publishers using AppNexus for header bidding, either by integrating with us as a demand source, leveraging our open source prebid.js wrapper, or both. We’re currently rolling out services to a select group of initial publisher clients.

In addition, will the account support mentioned in the release come at a premium?

Rubenstein: PreBid Enterprise is our comprehensive header bidding solution. It includes the same open source wrapper technology and demand source integrations that we launched a year ago, which continues to be free. The new services and analytics offerings have an additional cost. As a whole, PreBid Enterprise gives publishers the transparency of an open source wrapper with the option to work with a team of experts who can help drive business goals and provide technical implementation and ongoing support.

Header Bidding Generally:

The Drum: Header bidding was earmarked quite notably in the ad tech earnings calls last week, both specifically by investors, as well as by the trade press. Why do you think it has become such an issue whenever the technology has been around for years?

Rubenstein: Fundamentally, header bidding is a way to stop one company, Google, from limiting publishers to its own, proprietary demand. Header bidding allows publishers who use DFP to access all of their demand sources directly in a fair and transparent auction, to get the best price for their inventory. It’s been very disruptive to Google’s advertising business, as it threatens DoubleClick’s manufactured monopoly. What it’s done is allow digital publishers to earn much more money for their content, and other ad tech companies to bid directly for advertising inventory. While the technology has been around for quite some time, an increasing amount of publishers have adopted programmatic technology over the past few years, making the shortcomings of Google’s ad server clearer to the market. It’s only been more recently that the industry has realized that header bidding is good for everyone – except Google.

The Drum: What to your mind separates the AppNexus header bidding solution to alternatives that profess to be server-to-server side solutions, such as OpenX Meta, and PulsePoint's BidderExpress, etc.?

Rubenstein: As a technology company, we are always investigating different ways to help publishers realize the best market price for their inventory. We've seen dramatic success with our header bidding technology over the past year, and more recently we’ve prioritized building out services and analytics to best satisfy our clients’ header bidding needs now. There are still a lot of questions surrounding server-side header bidding, such as how it can differentiate from traditional RTB. It’s still early, and we don’t yet have a compelling reason to develop a server-side approach.

The Drum: Also, how would you say the release of Google's First Look, and EBDA has affected publisher take up of header bidding thus far? Would you agree that it has somewhat stalled uptake of the technology, or would you say it has had a negligible effect?

Rubenstein: One sure indication that Google is afraid of losing its monopoly is the introduction of First look and EBDA, or Exchange Bidding in Dynamic Allocation. It’s a clear effort to stall publisher adoption of header bidding, but neither product addresses the problem that header bidding is designed to solve.

At first glance, First Look seems a lot like header bidding. But instead, it establishes a preferred auction with a price floor. While a useful innovation in some regards, First Look does not resolve the structural limitations of DFP. Publishers are still unable to access multiple demand channels directly, and they are still locked into the black box that is AdX. In effect, First Look leaves you right back where you started.

Google’s EBDA program raises more questions than it answers. First, can we be sure that EBDA allows publishers to work directly with their chosen demand partners, and that those partners can bring the full heft of their demand? Second, does EBDA enable us to be fully transparent and honest with publishers? Google hasn’t answered.

The Drum: Are you able to offer any insights on how those publishers you've worked on the implementation of header bidding have seen when it comes to yield?

Rubenstein: Fundamentally, publishers are winning, and that’s why we’re seeing such great adoption. Through a fair and transparent auction of their inventory from many different demand sources, not just a select few such as AdX, they’re seeing revenue lift of 20 per cent, 30 per cent, 40 per cent and more.

The Drum: One of the Google arguments against the use of this technology would be its propensity to pageload lag. I've done some research, and there is a popular conception out there that each header partner requires a separate header tag, and that this can result in aforementioned lag, and ultimately risk spurring ad blocker take up. What would your response to this be?

Rubenstein: If publishers choose the right wrapper code and implement it efficiently, latency ceases to be an issue. Publishers should ensure their header bidding solution initiates an asynchronous auction, which means calls do not block the page, other header bidding partners, or the ad server from loading. If executed properly, publishers can increase monetization without degrading user experience. Additionally, with the right wrapper solution, publishers can set timeout limits to further prevent excessive latency.

The Drum: Finally, what would you say to those with concerns over data leakage, and how the implementation of a header tag may leak audience data to bidder partners? First of all, is it technically possible, and secondly, how do publishers arm themselves against this?

Rubenstein: Header bidding is no riskier in this regard than any other exchange-driven auction. If publishers know their partners and understand their practices and policies, they can safeguard their data.

The Drum approached Google for its response to AppNexus’ claims but it did not offer any further comment beyond those already published.

Although, a Google spokesperson did add: “Since launching the alpha, we are continuing to sign new publisher clients and new exchange partners. This isn't something Google is testing in a black box, it's in collaboration with the industry and our clients, taking their feedback into account.”

Ronan Shields

I'm the digital editor at The Drum, and cover adtech and martech. Prefer news and analysis, over opinion pieces. Current fascination(s) are blockchain and media futures trading; also curious about transhumanism on a personal basis. NYC-based, but really London Irish.

All by Ronan