Waterfall vs. Server-side or Client-side Header Bidding: What’s The Difference?

Updated on: January 5, 2024
While each has pros and cons, if you can make the technical side of things work, server-side and client-side header bidding will increase your yield.

Header Bidding, also known as pre-bidding, is an advanced method of integrating several demand partners (Ad Exchanges, SSPs, ad networks, etc.), collecting multiple bids from them at the same time, and selling the ad inventories at higher eCPMs unlocking higher yield.

It allows publishers to collect bids from demand sources for all of a publisher’s inventory before the sale. The goal is greater transparency to help publishers understand who is bidding for what amount. The technique relies on the users’ browsers for execution. 

In this article, we will understand the difference between client-side header bidding and server-side header bidding, the pros and cons of each type, and many more facts about them. So, shall we begin?

An Overview of Waterfall

In programmatic advertising, a waterfall model is a linear approach to selling ad inventory that involves sending ad requests to multiple demand sources in sequential order until an ad is served or all demand sources have been exhausted. The process starts when a user visits a website with ad inventories. The publisher’s ad server then sends a request to the first demand source, typically the highest-paying advertiser. 

If the demand source has an ad that matches the user’s profile and the publisher’s requirements, it will serve the ad. If the first demand source does not have an ad to serve, the request is sent to the next demand source in the waterfall, and the process repeats until an ad is served or all demand sources have been exhausted.

What was the challenge with the waterfall, and why was header bidding introduced? The waterfall method has limited reach, ad quality control, and yield management.

So, What Is Client-Side Header Bidding?

In client-side header bidding, bidding happens on the browser. Every time a page loads, ad requests are sent to the demand partners via a piece of JavaScript code added to the header of the publisher’s website. The demand partners conduct their auctions and send the bid prices received from the buyers to the page header.

Now, the code residing in the website’s header passes the highest bid value to the ad server (with the help of line item key-value targeting) to determine the final winning bid. Once the server receives the winning bid response from the header bidding line item, it compares it with the eligible server line items (Sponsored, Standard, etc.) and selects the line item with the highest CPM rate. Then, the server serves the ad to the user.

Pros of Client-side Header Bidding

Higher Cost Per Mille (CPM): Publishers can now receive bids from buyers who are more interested and are willing to buy more price.

Higher Fill Rates: More buyers mean higher chances of filling all available ad inventories, including premium and remnant (unsold) inventories.

Better Match Rates: Since the bidding happens in the browser, publishers can synchronize the cookies and identify the relevant users for advertisers. This capability of delivering targeted advertising offers more revenue. It also improves the user experience because users don’t see irrelevant ads.

Greater Control and Transparency: Publishers are the owner of their websites, and they are the ones managing the header bidding codes on the side. So, they can remove or add any demand partner, change the header bidding codes, and so on. Besides, it creates a transparent auction as the publishers know they are getting the highest bid from the buyers.

Cons of Client-side Header Bidding

Latency: Demand partners take time to get bid responses from the buyers. So, adding more demand partners means the JS code has to wait longer to receive their bids. And this leads the page to take more time to load, making users wait longer. It hurts the user experience, results in fewer impressions loading, and lowers the likelihood of ads even being viewed.

Compatibility: Compatibility with multiple browsers is important. Certain browsers may behave in certain conditions differently (e.g., some may pool connections to external pixels or block them completely).

Duplicate Bids: Risk of putting the same impression or inventory up for sale since a publisher uses multiple bidding partners. However, this isn’t limited to client-side header bidding; it can also happen with server-side header bidding.

What Is Server-Side Header Bidding?

It’s just like client-side header bidding, but the difference here is that instead of sending the requests from the browser, server-side header bidding sends them to the different demand partners from the ad server. This means the auction now happens on a server provided by a technology partner instead of the user’s browser.

There’s less implementation work because only one script (from the ad server) must be added for all partners. When a webpage loads, an ad request will be sent to the ad server. The ad server then sends the bid request to SSPs, DSPs, ad exchanges, etc., and receives the bids to determine the winner and serve the ad from the highest bidder.

Pros of Server-side Header Bidding

Reduced Latency: As the browser sends a single request (to the ad server) this time, the page load time is greatly reduced.

Scalable: The server-side approach essentially creates unlimited scalability without any price on latency. A publisher can add as many bidder partners as he/she wants. 

Cons of Server-side Header Bidding

Match Rates: The match rate will be lower since the ad request is sent to the SSPs, DSPs, and Ad Exchanges from an external ad server (not the user’s browser). Hence, bidding prices will also be lower than the client-side header bidding as the relevancy of ads will be diminished.

Black Box: Ad Server will be in control, and publishers must rely completely on the server to pass the right values and decide the winner. Publishers consider this as a ‘Black Box Approach’. On the client side, a publisher will be in control and can see what happens exactly.

Client-Side Header Bidding vs. Server-Side Header Bidding – Which One Is for You?

If user experience* is more important than CPM for you, then the ideal way is to go with server-side header bidding. If CPM is more important than user experience*, the right way is to go with client-side header bidding. However, latency can be reduced if you add the ideal number of demand partners (i.e., 5-6), then page load speed can be improved to a great extent via client-side header bidding.

*Note that we’re talking about very minimal differences here. For instance, when you execute and compare both solutions, there might be a few milliseconds difference. 

The technological advancement of header bidding is giving publishers options they never had before to monetize their inventory most optimally and efficiently possible. It has opened up a different realm of possibilities for publishers, but there are many factors to consider to make the best decision for your particular setup.

Still confused about which way to go? Hybrid header bidding is for those who don’t compromise on both. Some demand partners are shifted to the ad server in the hybrid header bidding. So, minimal latency and higher CPMs. A win-win condition.

News and Tips for Publishers

Get the inside scoop on publishing and programmatic with our 5-minute newsletter.