Header bidding has been an exciting development on the demand side. It unlocks many more premium supply deals, allowing for full-funnel customer management with clients and stimulates greater budgets flowing to publishers. Publishers also benefit from the additional control they get by being able to enable more 1:1, premium deals, programmatically.
So why change a good thing? What’s the benefit of moving to a server-side solution, which sounds complex and nascent in the current space? There are a few good reasons why it might make sense for publishers to take such a leap.
There are obvious efficiencies on both the demand and supply side from having a single, server-side solution in place. A DSP won’t have to submit the same bid to the same SSPs, all of which are then submitting final bids to the publisher, increasing costs substantially. And, even more importantly, it would create a unified auction for publishers... But is this 10+ years in the making?A unified marketplace would result in the highest yield for an impression, versus an unconsolidated second-price auction approach across multiple partners. It would also reduce browser-based header bidding page load times, increasing site views, customer satisfaction, and online sales.
There is even greater opportunity on both the demand and supply sides, if done properly through a publisher market consolidator. (PMC? – No, I won’t go there. Much too complicated a name, anyhow.) But - it requires a different type of company than the SSPs of today.The concerns with a server-side approach are losses in match rates, a lack of transparency on auction mechanisms that ensure fairness, and the requirement of SSPs to trust such a partner. While these are hard, they are solvable if the opportunity is large enough. Let’s look at each of these points in more detail:
- Transparency and Fairness of Auction Mechanisms– This is where a new type of entity would need to emerge, which would need to provide transparency in its auction mechanisms, as it would be imperative to the success of such an initiative. Publishers may feel a bit jaded about this (i.e., the promise of insights from exchanges have not fully come to fruition), or may only be concerned with the final CPM provided. But, publishers can and should think about how this ultimately benefits them and can work to demand it from such a partner.
In addition, browser-based header bidding, and using multiple partners, both add to the issue of lacking insights. SSPs don’t see or understand what is happening within the browser, and also can’t see what other partners are doing, whereas a server-side solution across all partners would make it easier to understand what exactly is happening in terms of bids and final yield.
- Match Rate Loss– Yes, it takes some time to increase match rates, but it is achievable. We have match rates upwards of 90% with many partners today. But, the real opportunity here is to leapfrog cookie-based match rates. We should match based on people rather than devices. By partnering across DSPs, a single auction consolidator, and SSPs, we could as an industry provide a real, open alternative to the closed marketplaces and walled gardens that depend heavily on their proprietary user graphs to show value to advertisers.
- Will SSPs Get on Board?– This might be the highest bar as it requires a sort of co-opetition. The challenge can be obviated somewhat if an entity is truly acting as a pure, consolidated pipe for a unified auction (think: tech fee). If it is truly a better solution for publishers (and there is a player that is truly transparent around their auctions and hence qualified to be an objective bystander) SSPs may have to work very hard to offer a better solution.