On paper, header bidding provides a perfect solution to the inefficiencies of the ad-serving waterfall.
Also known as pre-bidding or advanced bidding, the technology allows publishers to offer impressions to multiple ad exchanges simultaneously, before their primary ad server is called. This avoids the pitfalls of the one-at-a-time waterfall model and allows publishers to maximize yields while fully understanding the value of their inventory.
But header bidding does have some much-publicized drawbacks. In addition to scripts being complex for developers to install, the synchronized call-outs to multiple ad exchanges frequently cause unwelcome latency issues that slow down the publisher site and impact the user experience. Like many solutions developed to automate digital advertising and increase efficiency, header bidding still requires some refinement and improvement if it is to become a permanent fixture in the digital advertising landscape.
So how has header bidding evolved to date and where will the technology go from here?
Header bidding was originally designed as a temporary fix to a specific issue with Google's DoubleClick ad server. Google's dynamic allocation didn't allow bids to be submitted for individual impressions to compete with its ad exchange, instead estimating what price an outside exchange might bring in. The introduction of header bidding bypassed this situation, providing publishers with actual bids rather than just estimates.
But Google has now launched a new feature in the form of dynamic ad serving. This latest functionality allows multiple ad exchanges to submit real-time bids using server-to-server communication. Although there are a few teething issues with this new feature -- such as increased technology costs for buyers and the possibility of them bidding against themselves -- dynamic ad serving essentially provides a shortcut in the process of header bidding, leading to suggestions it may quickly become redundant.
Despite Google's latest rollout, header bidding may still have potential. Even in its current unfinished state it provides a better model than the one publishers relied on previously, and with further development and refinement it could become an exceptionally powerful tool.
Here are three potential pathways header bidding may take from here.
The first, and perhaps most likely future for header bidding lies in the hands of the corporate giants. Over the coming months they may well gain the upper hand in header bidding and -- as with Google's dynamic ad serving -- release their own solutions to the issues surrounding the efficiency of automated advertising, launching ever-improving features to assist publishers in multiple-bidding scenarios. The probability of header bidding taking this path in the future will be greatly increased if Google's dynamic ad serving -- still a relatively new technology – proves to be a success.
Recovery and reinvention
The second option is header bidding may be taken on by third-party developers who could re-master the technology to resolve its inherent shortcomings, resulting in a solution that surpasses both header bidding in its current state and Google's new offering. This option is most likely to come to fruition if Google's dynamic ad serving turns out to be just as problematic as header bidding.
The final possibility is that header bidding in its present form will fizzle out entirely. This scenario is probable in the face of an as-yet unknown technology emerging that overcomes the weaknesses of header bidding, making the concept obsolete and providing a dominant new solution.
It is too early to be sure which of these futures beckons for header bidding; much hinges on the effectiveness of Google's dynamic ad serving. What is certain is that -- despite its disadvantages -- the invention of header bidding technology has allowed automated advertising to take a significant step forward, helping publishers to maximize yields and getting the best ads in front of consumers.