BlogHeader Bidding

Server Side Header Bidding Explained

What is Server Side Header Bidding?

Server aspect header bidding guarantees to enhance latency, scalability, and auction logic issues seen in traditional header bidding by shifting communication with exchanges away from the browser and into servers.  The process is actually the identical method SSPs and DSPs integrated for years, except this time the SSPs are integrating with one another.

You’ll be able to see the distinction in process within the diagrams under.

For my part, server aspect header bidding has been sluggish to see adoption for a number of causes but is poised to take off in 2017.

Publishers Will Demand It

Most publishers saw massive increases in programmatic revenues over the previous yr by shifting to a header bidding setup.  But those features have been as a lot pushed by adding new companions as they have been by header bidding tech itself.  The extra partners publishers labored with, the more unique demand they have been capable of seize.  Now, nevertheless, many publishers are starting to run into technical limits on the browser aspect which forestall them from adding new corporations within the header.

This restrict is the utmost concurrent connections a browser is designed to deal with directly.  Chrome for example will only make six requests per host and ten general earlier than it waits to situation new requests.  Principally the browser renders the page in batches of ten requests at a time, so it doesn’t matter for those who put fifteen header bidders in an asynchronous wrapper if the browser can solely call ten directly.  So for publishers, server aspect header bidding seems like an incredible answer so as to add as many companions as they like without having to fret about paying a worth on consumer experience.

Know-how Suppliers Gained’t Have a Selection

So far, header bidding has been a unbelievable catalyst on the gross sales aspect for know-how suppliers.  They not had to outperform their entrenched competitors at a publisher or maneuver for waterfall place. Any fill was incremental fill in a header bidding world, and publishers have been eager so as to add new demand to their networks. Just like the publisher predicament nevertheless, as a result of most clients have already crammed their obtainable positions in the header, know-how suppliers are back to having to earn their relationship.  Distributors need to return to a world the place they will still sell their providers as “no cost” to the writer they usually can have it by means of a server aspect level of entry.

Finally, there are some publishers who have been sitting on the sidelines with header bidding as a result of they see it as a a basically inefficient idea.   These could possibly be huge, brand identify publishers who run advertising as a aspect business.  Assume huge e-Commerce gamers who are laser targeted on how pageload impacts checkout fee.  These media corporations are the last whales left for most of the know-how suppliers, and a server aspect strategy puts them back in play for sales organizations.

How Server Side Header Bidding Works for Publishers

Server aspect header bidding works primarily the same method as common header bidding from the publisher perspective, there’s simply less implementation work to do.  A publisher still has to put some JavaScript code in their header with a view to set off a request to an change earlier than their advert name, but as an alternative of 1 script per associate, there’s just one script general that’s needed.

On the advert server aspect, there’s really no change; publishers can determine to setup one set of line gadgets across all companions, or a set of line gadgets per companion.  Nevertheless, because there are sometimes tons of to hundreds of line gadgets required per companion and as mentioned before, a key benefit with server aspect header bidding is with the ability to add many more companions, it’s extra probably for publishers to select the previous choice and consolidate the ad server setup into a single set.

How Server Side Header Bidding Works for Exchanges

The actual distinction is on the trade aspect, particularly for the change that has the ‘tag-on-page’ place and is the platform that calls out to all other platforms server aspect.  This might sound confusing, but with server aspect header bidding, at the least one platform has to maintain a standard or shopper aspect setup with the writer.  Solely AdX maintains a 100% server aspect reference to the ad server, although it does supply third social gathering exchanges a degree of server aspect access by means of its Change Bidding Dynamic Allocation (EBDA) product.

For the server with tag on page nevertheless, additionally they have to handle integrations with all other SSPs the publisher needs bidding on their stock.  This integration works very similar to the OpenRTB integration between SSPs and DSPs in that bidders (whether or not they are DSPs or server aspect SSPs) submit their true bid and not the shut worth of their auction.  This enables the tag on page trade to run a true, unified public sale across all server aspect connections on behalf of the publisher.

Why a Unified Public sale Matters

One of the key advantages to header bidding is that it allows publishers access to extra demand; the issue is that with the normal header bidding setup, the demand remains fragmented in every trade. This creates bizarre situations where the very best bid doesn’t all the time win the auction.  Take the example under:

Here, Trade A has bids of $5.00 and $1.50, whereas Change B has bids of $3.00 and $2.50.  If they each conduct a 2nd worth auction, Change A will submit a profitable worth of $1.51 to the writer, whereas Trade B will submit a profitable worth of $2.51 to the writer.  Trade B wins as a result of it had a better second bid.  This is exactly what occurs at present in the header bidder setups most publishers are utilizing.

Non Unified Auction Header BiddingThis result’s great for Change B, OK for the writer, and dangerous for the $5.00 bidder.  Clearly the perfect the $5.00 bidder in Change A would win the impression, since they have been the very best bidder, and pay $3.01, using Change B’s bid in the identical, unified auction.  Server aspect header bidding allows this ideally suited state of affairs because there is a central platform (vs. just a static JavaScript wrapper) that can see all bids, handle the auction logic, and notify winners of their worth paid.  Every change is already built to do this inside its own system with DSP bids, and different SSP bids aren’t any totally different.

Unified Auction Header Bidding

So more good news for publishers is that server aspect header bidding has the potential to enhance revenues with the exact same companions already in place merely by means of this unified auction.  It’s the identical influence conventional header bidding had on AdX trade by change; now those features can happen collectively and across any change.

So Does Server Side Header Bidding Scale back Latency?

The answer right here is yes, completely it does – but the place you actually see the positive aspects is on the scalability aspect.  This stuff are intertwined; header bidding isn’t scalable beyond six to eight partners due to the browser’s concurrent connection restrict, and the issue that limit creates is latency.  The server aspect strategy primarily creates limitless scalability without any worth on latency.  Not solely are you able to scale back the latency you have got now, however you’ll be able to add as many companions as you need server aspect without worrying about impacting pageload once more.

About Those Cookies, Though

Sadly, server aspect header bidding isn’t all good news.  The most important downside to this new know-how is that it has the potential to hurt monetization by means of poor cookie match charges.  This occurs because the SSPs that transfer from a shopper aspect setup to a server aspect setup lose access to the top consumer and may’t assign or learn their cookie any longer.  With conventional header bidding every consumer calls each trade on each web page, which could be lousy from a latency perspective but is incredible from a consumer sync perspective.  Each change can always and instantly determine each consumer’s nameless cookie.

When exchanges transfer to server aspect they should determine every consumer by holding their cookie in sync with the tag on page trade’s cookie.  This provides a center man to the equation and makes the server aspect exchanges solely depending on the tag on web page change syncing with them typically.  An unsynced customers is identical as a cookie-less consumer, and sometimes nugatory to an trade.  There’s no knowledge hooked up to the consumer, so bidders can’t worth it at rather more than a blind prospect, which isn’t value much.  Unsynced users are more likely to get extraordinarily low bids or no bid at all, so the impression is critical.

How much worse sync charges are server aspect vs. shopper aspect is tough to say, because it will depend on who the two parties are.  But, protected to say that the larger the change, the higher their sync price is more likely to be with each platform, so publishers can be sensible to anoint a serious SSP to their tag on page place vs. a smaller platform.

Strategic Outlook

As you’ll be able to in all probability guess, the tag on page SSP has some major benefits over the server aspect SSPs.  Because their code sits on the publisher’s website, they have a a lot stickier relationship with the top buyer, to not mention they’re the only platform that maintains access to the top consumer (and their cookie), in order that they’re more likely to monetize better.

Talking of monetization, due to the cookie sync situation it’s not clear to me if server aspect header bidding will truly make publishers extra money than conventional header bidding.  Positive you’ll be able to add extra partners and save on latency, however those will increase might not pay for the losses incurred by fewer data-rich customers.  I might not take it as a assured that server aspect header bidding will make publishers extra money, as a lot as SSPs may determine to position it that method.

For my part, the bigger strategic impression of server to server header bidding is a continued push towards complete commoditization of supply and what the buy aspect response to that could possibly be.  In distinction to the waterfall where DSPs needed to take heed to all SSPs as a result of all of them had entry to totally different stock, header bidding has shifted the business to a world the place every platform has access to all supply.  That makes the exchanges much less distinctive and fewer worthwhile to DSPs. This duplicative provide burdens DSPs without offering any incremental worth and that can’t final endlessly.

If consumers start pulling demand from smaller platforms and consolidating onto giant marketplaces although, it’ll remove the necessity for therefore many companions within the first place.

Related posts
accessibilityBlogbrowserChromeedgeFirefoxInternet ExplorersafariusabilityUXWeb

Group Labels Do Not Guarantee… Uniquity?

AdoreboardanalyticsBlogcustomer centricitycustomer experienceCXdata scienceemotion analysismarginal gains

Are Marginal Gains the Answer to Measuring Customer Experience? – Adoreboard Blog


Old Forge’s new passion: mountain biking


David Gibson On APA Appointments; Role of Statewide Interests –