One of the most effective ways for publishers to drive up CPM rates is to ensure that enough good demand partners are bidding on their inventory. There are a number of ways to do this, each with it’s own benefits and drawbacks, but publishers are increasingly turning to Server to Server bidding, AKA S2S Header Bidding.
For much of the history of programmatic advertising this was largely achieved through ad waterfalls, which did the job but were far from efficient as only the SSP at the top of the chain gets the opportunity to bid on every impression. More recently Header Bidding has become the favoured approach.
How Server to Server Bidding Works
“Server to Server” refers to the ad-server and the demand partner’s server communicating directly. When an ad request is made, the ad-server connects directly to each SSP to get their bid. Because this process happens in well connected data-centres rather than in the user’s browser, this can happen very quickly.
S2S benefits from faster processing, being able to request and receive more bids at once and in being able to transfer information far faster than most users connections would allow. The benefits of this are two fold: firstly, it happens very quickly, secondly there is no additional load or delay for adding more exchanges to the process.
How to implement Server to Server Bidding
Large publishers looking for a Server to Server solution have the option of developing their own wrapper, or building on a solution such as Prebid Server. This is not a simple undertaking though, either in terms of the technical challenge or even in getting SSPs involved. For most publishers, S2S means implementing either Google’s Exchange Bidding or Amazon’s Transparent Ad Marketplace.
Header Bidding vs Server to Server Bidding
Server to Server bidding is often seen as an alternative to Header Bidding. This isn’t always the best way to utilise S2S, but it is useful in terms of understanding the differences. The main advantage that Server to Server bidding offers is speed. Delivering ads to the user’s screen more quickly not only increases revenue by getting more ads seen, but it improves user experience and therefore encourages users to venture further into the website. This can increase impressions per session even further. However, S2S bidding is not without it’s drawbacks.
The biggest issue with S2S Bidding is the problem of Cookie matching. Advertisers bid more when they know more about the user behind the impression that they are bidding on. This usually translates to the user having information in a cookie (such as that they recently visited the advertiser’s store and viewed an expensive product). With client-side Header Bidding, this is simple as the auction runs in the browser, each bidder can look for any of “their” cookies on the user’s system. S2S bidding can’t leverage these cookies as it runs on the server not in the browser, and it is far more difficult to find that targeting information.
The advantage of a hybrid approach
The best solution at this time seems to be a hybrid approach. As S2S bidding can be run without additional latency, it can be added into a Header Bidding set-up without impacting user experience. This can allow the Header Bidding set-up to be slimmed down, with high performers remaining, and S2S bidders added to bring more bids at high speed.
This approach is one of our favourite approaches when managing website monetization for our clients. We manage ad inventory on behalf of publishers, introducing top performing SSPs through Header Bidding, Exchange Bidding and non-banner formats. Because we manage both the technical set-up and the day to day management, our publishers are free to focus on developing their business whilst we increase their programmatic ad revenue. Find out more about how we help publishers here.