Header Bidding vs Open Bidding – What Should You Know!

Two pillars of programmatic advertising that marketers still struggle to differentiate between: Open bidding and Header bidding. Who’s the winner?

Header bidding severed the long-standing dependency on Google for ad monetization with flexibility and level-play, which is why Google gave a counterattack with Open bidding. 

Both technologies serve the same purpose and efficiency, which is the same reason that makes it hard to differentiate and understand each to choose the right one. This blog gives answers to these very perplexions and explains the significant differences. By the end, you will know the way to choose the right technique that gives high CPMs and ad revenue.  

So, let’s get started.

But before learning about differences, let’s understand what do they mean 

What Is Header Bidding?

Header bidding or pre-bidding is an advanced programmatic technology where you can open your premium inventories to multiple demand partners (both guaranteed and non-guaranteed line items) simultaneously to bid and place ads. 

Header bidding can be done both on the client side and on the server side, where it sends requests directly from the website to demand partners or from the ad server to demand partners. 

What Is Open Bidding?

Google’s Open Bidding is a server-side unified auction where several ad exchanges and SSPs compete with Google’s Ad Exchange (AdX) to win the impressions, just like header bidding.  In open bidding, one request from the site to Google Ad Manager (GAM), the ad server, is enough. Further, the GAM will take over the process of finding the highest bid among Google AdX and non-Google ad partners. 

What’s the Difference Between Header Bidding and Open Bidding?

The first and main difference between header and open bidding is the auction process; let's take a brief look at the differences between each process: 

Header Bidding Process

As mentioned before, in the header bidding process, ad requests from your ad server are sent to several demand partners simultaneously. Each time an ad request is sent, the header tag initiates the auction, collects bids from all demand partners, and then passes the highest bid to the ad server. 

Now, the ad server compares the highest bid with sponsored, direct, or programmatic deals and serves the ad creative to the user. Since the received bids are made to compete with the ad server’s demand, it substantially increases your eCPM and revenue.

Open Bidding Process

Google’s Open Bidding is a server-side unified auction where several ad exchanges and SSPs can compete with Google’s Ad Exchange (AdX) to win impressions, just like header bidding. The only difference is that the auction occurs on the server rather than the client. 

To be frank, Open Bidding was a counter punch by Google to Header Bidding vendors. Google even dropped its last-look advantage in Open Bidding to make itself more friendly.

The main difference is that the auction will happen inside GAM (ad server), not on the users’ browser. Here’s what the process looks like:

• An ad request is triggered, and the information is passed to the GAM ad server.
• GAM runs a unified auction to determine the best yield
• GAM selects the best-trafficked line item to compete in the unified auction.
• It sends a bid request to targeted yield partners.
• Targeted yield partners run their auction and return their most competitive bid to GAM.
• GAM hosts a unified auction and selects a winner.
• A creative is served to the user.

Source

Besides the difference in processes, their features, pros, and cons widely vary

Here’s a quick overview of it with the comparison table.

Which One Is Right for You?

Both have their pros and cons. For instance, header bidding will be completely transparent and can be managed or customized by publishers. And, with the availability of open-source header bidding technologies like Prebid, it has become much easier to implement and execute header bidding.

On the other hand, Open Bidding helps you to deal with page latency. As we’ve shifted the auction to the server side, the number of ad requests and wait time can be reduced. But the server takes care of the auction, and publishers have no control here. 

Both techniques may differ in latency and control attributes, but they possess high efficiency and auction performance. 

Therefore, it depends on your needs whether you are going to choose high demand, efficiency, and control or high demand, efficiency, and low latency. You can find the right fit for you in two different cases: Choose either one with testing or Choose both.

Choose any one

It doesn’t hurt to test and try each technique separately for a stipulated period and compare the outputs. This way, you can find the right one with the quantitative results of your website rather than relying on documented numbers.

Choose both

What I mean by choosing both is to implement hybrid header bidding. Through this technique, you get to use both client-side and server-side open bidding. Thereby, you can increase the demand and revenue. 

In both cases, choosing hybrid header bidding might seem like a great choice. But again, the great choice stays so if it suits your revenue goal and website technicalities.

So, overall, the strategy lies in testing combinations and comparing results. Eventually, you will be driven to the right pick that gives high impression values and overall ad revenue.  

Header bidding and Open Bidding Best Practices

Auction efficiency relies on many parameters that have to be essentially optimized: floor price, bidder count, and bidder timeout. 

• Optimizing floor prices is essential to get a fair enough value irrespective of changing demands and seasonality. If you are using header bidding, you can use dynamic floor pricing tools to adjust the Prebid floor prices in real-time and achieve higher revenue. 

• In accordance with open bidding, Google itself offers unified pricing features, which you can use to add rules on how to change floors according to the changing demand. This way, you reap the revenue that you deserve on changing demands. 

• Checking the bidder count frequently is vital to remove and replace the idle ones to reduce latency or give a chance to high-value bidders who are affected by the latency caused. The bidder count might change from website to website, but utilizing ten bidders is the safe limit proposed by the Prebid experiment. 

• Lower timeouts prevent all bidders from sending back responses, and higher timeouts cause latency. So, optimizing bidder timeout to balance these mismatches is essential to maintain your revenue potential. Use timeout tools to optimize timeouts and install proper timeout measurement systems to help you find the source that causes delay and rectify it. 

Final Thoughts

Ultimately, these two bidding solutions offer the same advantages: bidding opportunities for the demand partners, better CPMs, and ad revenue for the publisher. 

There are pros and cons to both Open Bidding and header bidding, so the right solution will ultimately depend on what you’re looking for. If you can’t seem to decide between the two, consider them both as options, as they can ultimately coexist within your ad server. 

Fortunately, you’re never obligated to choose one over the other—make the call that works best for your setup, even if it’s called something different. You can always change course in the future if you find yourself wishing you had picked a different solution from the get-go.