#programmatic

The tech is not new. But almost 3 year’s on, Amazon’s transparent ad marketplace  is changing the header bidding game slowly. Among the most popular server side products  in the ad industry,  server side bidding is slowly solving the latency issues browser side header bidding has faced, with slow page load times (where ad calls would be made on people’s browsers) by moving it to large servers hosted on the publisher’s end. 

And how many players can compete with Amazon server capacity?

Innovation in ad tech is ongoing. When header bidding came about three years ago, it kind of changed the bidding approach in automated advertising. Publishers could simultaneously offer inventory to multiple ad exchanges, before making calls to their ad servers. This changed the waterfall approach where bids were passed sequentially from one exchange to the other in the earlier days.

It also yielded better results. Per Digiday, 58% of users reported higher CPMs, 23% reported fewer passbacks and 31% reported higher yields. 

Great numbers in a playing field where ROI margins can be tight.

Now the slow but “gradually growing” new trend is server-side bidding (for Headers)  and a long term view is that  it will slowly replace browser side bidding completely.  

But there is a caveat.

With server-to-server wrappers, one vendor aggregates the bids from all the other vendors in a cloud environment. This means only the vendor collecting the bids has access to the user’s browser. Read monopoly and an unfair playing field for other publishers who are also screaming for attention and ad dollars.

This also means that other vendors have to sync their data to the aggregator’s data. 

This additional step can reduce match rates and give lower numbers of custom audiences mapped to CRM or pixel data. It is one reason why publishers are hesitant to adopt server-side bidding, since it can lower advertiser spends or hedge everything with one player.

With Google and Amazon, the two tech titans truly in the server side game, this is a bit more balanced because both of them have a lot of personal user data. Google has search data which defines intent, while Amazon has commerce data (which gives very clear indicators of people’s buying habits).

Google’s product is exchange bidding and it’s just rolling out (layered with all the data from its search, gmail, youtube usage) - and the only company which can realistically compete with it will be Amazon.

<Tangent/>

Facebook is also in the game but currently partnering Amazon, AppNexus, Index, Sonobi, Sortable and Media.net and also partnering PubMatic, AppNexus and Index in an open source prebid server side project titled prebid.js. 

This means that any publisher that has PreBid.js installed on their website can now access the new software that has server-to-server support built in, and then choose which header bidding partners they want to transact with, thus increasing their access to buyer-demand.

The fightback is clearly on. 

<Tangent/> ends.

It remains to be seen, how fast Amazon scales its ad (as well as ad-tech) offerings. It goes unnoticed in media which is mostly fixated on Jeff Bezos and AWS, but in less than a year,  21% the top 1000 websites in the US are using a hybrid mix of server side and browser side wrappers. 

Amazon (40.3%) is already 3rd in the list of bidders behind (Index 48% and AppNexus 47.9% - source serverbid)

With Google coming in with a dedicated cloud ad-tech (server-side) product, server-side bidding will have more takers and while advertisers are for now using a mix of browser and server side bidding, given Amazon and Google’s scale (and user data) , the question is how long before it all goes in server side, especially if benchmarks improve?

What would be the publisher distribution approach as well as revenue split then and who will actually control it?

Daily Active Users (DAU’s) dictate scale which dictate data which dictate aggregation and distribution power. 

Monopoly much?

Nah. Just business.