Header bidding has been one of the key issues of debate in the sphere of programmatic advertising during the last 12 months, with publishers asked to make a choice between this, and the Google alternative. In this piece, Andrew Buckman, managing director, EMEA, at OpenX, outlines some of the key issues of consideration in the debate.
Ever since it hit the digital advertising ecosystem, the whirlwind of header bidding has not lost an iota of momentum. Over half of large publishers are already using the technology and it is proving to be a force so powerful that even Google couldn’t stop it.
Indeed, if Google’s decision to open up Double Click for Publishers (DFP) to outside demand shows anything, it is that header bidding is a long term play and not a passing fad.
The opportunities it offers for both buyers and sellers are increasingly gaining recognition. While publishers are entranced by its ability to drive yield higher than previous solutions with dynamic pricing, buyers are also beginning to discover the appeal of its enhanced transparency and early access to premium inventory.
The only question now is: what should we expect for the future of header bidding? But before we get to what’s next, let’s recap on where we have been.
How it works and why it’s important
Header bidding is an automated trading tool that enables publishers to instantly offer inventory to multiple demand sources, prior to calling the ad server. The technology functions by placing a piece of code within the header of a publisher’s page that can call a demand partner — such as an exchange or private market place (PMPs) — and relay the results of a real-time auction, all before the ad server has loaded the first level of the traditional waterfall approach. Once the highest bid is selected, it is sent on to the ad server and the chosen creative is displayed.
In the traditional programmatic environment, buyer access was limited according to order of preference and terms set by the ad server, including price, with little visibility into the audiences that impression would reach. Not so in header bidding. By making inventory available to all demand sources, it not only drives competition and yield — sending the best bid from the auction to the ad server — it also allows buyers to get a detailed first-look at inventory and choose only the impressions they want.
The early issues of header bidding
No new technology is without its shortcomings and header bidding is no exception. For the most part, issues have centred on demand sources — both in terms of the time it takes publishers to integrate them, and the latency problems that arise when several are in play. The desire to connect with as many demand sources as possible can mean publishers run too many header tags at once. Consequently, they are not only difficult to manage but can also slow down ad load times — meaning users consume content before ads can appear.
This has a knock on effect for viewability, as ads that take too long to load are not seen.
Containers and the streamlined fix
One solution was quick to emerge: header bidding containers. The technology wraps multiple header bidding tags into one big tag that is less cumbersome to use, inserts into a page header, and manages the bids submitted by the bidders. It addresses some concerns by reducing the stress of implementation by streamlining the way demand partners are added and removed.
Even though it has been around a short time, the tool is already evolving – moving from in-browser solutions that suffered from latency caused by container weight or inadequate network speeds to server-side container solutions that shift trading onto the server. In doing so, it frees up user browsers, speeds up the process of ad calls, and increases stability.
A broad horizon and a bright future
So is Google’s dynamic allocation enough to overtake header bidding? Not by a long shot. Aside from the fact that its solution does not support vital elements, such as PMPs and dealID; one of its key branches — DFP First-Look — currently only includes demand from Google Ad Exchange, meaning there are significant portions of its ad trading system not open to external exchanges.
With its founding blocks of fair prices, open access and transparency; header bidding is expected to soon become a widely used tool for video and mobile inventory. Notoriously scarce, and spread across a disjointed marketplace, video and mobile impressions are often not exposed to the majority of programmatic demand. Header bidding will help to bring more buyers and sellers together, giving buyers the opportunity to utilise premium formats and enabling publishers to achieve maximum value for impressions rather than trading in bulk with a select few.
When it comes to programmatic direct, its potential is even greater. Currently, the attraction of trading via programmatic direct is that publishers have complete control over chosen demand and buyers receive better quality, often data-enriched, impressions. Yet restricted inventory volumes can mean the number of impressions a campaign delivers is low. This is where header bidding can change things. By making all inventory available to direct buyers, publishers can then prioritise demand sources at scale, thereby enhancing both delivery and revenue.
The header bidding whirlwind has blazed quite a trail and with so many new areas to shake up, it is not done yet. In the future, the technology will help publishers go beyond the banner and offer a range of high-impact ad units, as well as giving buyers access to premium inventory on an unprecedented scale. Destined to transform ad trading as we know it, header bidding is not dead; it is only just coming to life.