Header Bidding and the Programmatic Ecosystem

Header Bidding and the Programmatic Ecosystem

Header Bidding is the latest frenzy to consume the media and advertising industry.  Publishers are toiling away in labs, trying to figure out how to take back more of the revenue pie, while advertisers lie awake at night fearing increased costs and changes to inventory access.  Like many industry changes, header bidding started as a low hum, and has now moved into the over-hyped, can’t-live-without-it, will-solve-all-our problems phase.  Next will be disillusionment, then acceptance, and finally it will become the new normal.  While we wait for this process to push forward, let’s play catch-up on where we stand.

A Little Background

Publishers and exchanges have been concerned that the traditional means of auctioning off impressions through the programmatic buying process have been unfair.  Currently, publishers with inventory to sell rely on “waterfall bidding” – sequentially placing the exchanges most likely to bid the highest price near the top of the waterfall, and the least likely near the bottom of the waterfall.  With this process, ad calls are made to one exchange at a time, moving down the waterfall if the impression is not sold through the previous exchange.  Exchanges positioned lower in the waterfall could potentially lose an auction despite having an advertiser willing to pay more for an impression.

In layman’s terms – imagine trying to sell your house and receiving three offers.  The first offer is below list price, so you ignore it and move on to the second offer.  The second offer is slightly above your asking price, so you go ahead and agree to terms with the second buyer, without even seeing what the third buyer had to offer.  Who knows, it could’ve been more!

So What is Header Bidding?

Header bidding gives publishers an opportunity to get away from the waterfall.  Publishers add code into the header of the webpage (hence the name) which allows multiple exchanges to submit their bids at the same time, rather than sequentially, allowing the ad impression to be rewarded to the true highest bidder.  The result is a more balanced playing field where a publisher can get highest market value for the ad space on their webpage.

Back to our home-selling analogy: now imagine you’re sitting down to listen to offers for your house and four potential buyers all slide their offer across the table to you simultaneously.  You now have the chance to see each offer before you ultimately make your decision.  There’s no risk of missing out on a bid, and you will know that you got the best possible deal.

So This Is Bad for Advertisers?

Not necessarily.  While the promise of header bidding is higher prices for publishers, as programmatic advertisers, header bidding will allow us to gain access to high quality inventory that may not have been previously available.  This can extend the reach and impact of any campaign.  If the price is not right, then we can still pass on that impression – but at least we have more of a chance to buy it if it is valuable.

What are the Challenges with Moving Forward?

The header code required is robust, requiring some heavy lifting from the publisher’s ad operations team to implement across each of their webpages.  To maximize earnings, potential publishers must place code from multiple exchanges into the header, further multiplying the work required.  Additionally, the amount of code required has the potential to slow down the page load, creating a bad user experience.  This poor user experience could lead some users to adopt ad blocking technology, or choose to avoid the site altogether.

So far, header bidding has only been focused around desktop advertising.  Slow page loads and data fees greatly impact what can be done in the mobile web space.  The mobile app environment is experimenting with header bidding, but the technology would have to be more integrated into each app’s Software Development Kit (SDK).  Manipulating the SDK requires apps to resubmit to the app store for approval, adding another layer of complication to the process.

A Light at the End of the Tunnel

Google has modified their popular ad server DoubleClick for Publishers (DFP) to allow all bidders to compete evenly, including against Google’s own exchange, AdX. This has the potential to level the playing field without the implementation of additional header code.  If all exchanges have a seat at the auction through DFP, then perhaps this need for additional code will go away.

What isn’t going away – regardless of the solution – is the move toward a fair auction.  That means that publishers will be more willing to automate sales of their most premium inventory and sponsorships, all of which would be available to the highest bidder. However, custom content and premium video will likely remain in the hands of the publisher, not open to auction.

When One Door Closes…

Advertisers and agencies should continue to monitor header bidding developments and the impact this will have on premium inventory access.  Right now, the impact on the day-to-day management of programmatic campaigns has been minimal.  However, as the access to premium inventory expands, the true promise of programmatic buying may be fully realized.  For the first time, programmatic media buyers will be able to access almost all available inventory, further enabling them to deliver the right impression for the right audience.  Buyers will be able to scale delivery against multiple premium sites and use real-time decisioning to put ads in front of the most responsive audiences.

It’s important to know that while this change will absolutely happen, the current state of header bidding is far from a finished product.  Most publishers seem content to wait until all the bugs are worked out before diving in.  Eventually, header bidding may have a profound impact on the way we purchase digital media.  Many more aspects (technological and otherwise) still need ironing out, but when all is said and done, header bidding shouldn’t be keeping you up at night.

No Comments

Post a Comment