Header bidding lets publishers bring more competing demand to their inventory and increase their ad revenue. Prebid is one of the most widely used header bidding solutions in the industry, and its auctions can run in the user’s browser, on a server or as a combination of both. The choice of setup affects both ad revenue and site performance. Restrictions in the browser environment have also increased interest in server-side solutions.
There is, however, no one-size-fits-all setup. In this article, we look at the benefits and challenges of Prebid Server, compare building it in-house with using a technology partner’s solution, and share what our own testing reveals about the performance and revenue potential of different setups.
What Is Prebid Server?
Prebid Server is an open-source solution maintained by Prebid.org that lets you run header bidding auctions on a server (server-side) instead of in the user’s browser. It receives an auction request, forwards it to the configured demand partners and collects the bids. The response is returned once the bids arrive or the auction’s timeout is reached.
In a traditional client-side setup, the Prebid.js JavaScript library collects bids directly from demand partners in the browser. With Prebid Server, Prebid.js can instead send a single request to the server, which handles the calls to the different partners. This reduces the load on the browser and can improve site performance. However, loading and rendering the ads still happens on the user’s device.
What Are the Benefits of Prebid Server?
One of the most significant benefits of Prebid Server is better site performance. When auction-related processing moves from the browser to a server, page load times can get shorter, especially on slower devices. In our own testing, too, the server-side setup delivered the fastest page loads and improved ad viewability, especially on desktop.
Another key advantage is that you can grow demand without the browser load growing at the same rate. Testing and onboarding new demand partners is more flexible, as each partner doesn’t need its own browser-side integration.
Prebid Server also makes it possible to serve several advertising environments through the same infrastructure. Beyond websites, it can be used in mobile apps and digital out-of-home (DOOH) advertising, for example. Centralized management makes things easier, especially when a publisher runs multiple sites or advertising channels.
In addition, a server-side setup reduces the need for third-party code in the browser and gives publishers more control over the data passed to partners. It also makes the setup less dependent on changes in the browser environment.
Prebid Server for Mobile Apps: Prebid Mobile
In mobile apps, Prebid Server is an essential part of the Prebid Mobile solution. The SDK integrated into the app communicates with Prebid Server, where the auction takes place. This requires the publisher to have a Prebid Server environment, either its own or one maintained by a partner. Relevant Digital has been monetizing mobile apps with Prebid Server since 2020.
Through Relevant Yield, we have helped app publishers grow their ad revenue and optimize their header bidding setups. German weather service WetterOnline, for example, grew its ad revenue by more than 10% with its Prebid Mobile integration. Results also depend on the choice of demand partners, auction settings and ongoing optimization.
What Should You Consider When Using Prebid Server?
One of the key challenges of using Prebid Server relates to user identifiers and privacy signals. In a server-side auction, demand partners may not have access to the same identifiers and other signals as in a client-side setup. This can affect both the number and the value of bids. Remember that user consent and privacy requirements must be respected in a server-side setup too.
Technical compatibility is also worth checking before you go live. Not all demand partners, Prebid features and reporting options necessarily work the same way server-side as they do client-side. That is why the setup should be planned around the publisher’s ad tech stack and business needs.
The third factor to consider is the balance between costs and revenue. Server capacity, maintenance, and service fees add costs that can grow with the number of auction requests. An increase in auctions or winning bids alone therefore says little about whether the setup pays off.
Beyond net revenue, assess results with metrics such as revenue per ad request, fill rate, response times, and rendered impressions. This gives a more complete picture of how the chosen setup affects overall ad revenue and efficiency.
Prebid Server Implementation: Build In-House or Use a Partner?
Prebid Server can be run either in-house or as a service provided by a technology partner. The best option depends on the publisher’s technical resources, specific needs and how much responsibility for maintaining the infrastructure it wants to take on.
An in-house setup gives the publisher direct control over the server environment and the ability to tailor the solution to its own needs. However, it requires ongoing technical expertise and resources. Prebid Server’s official hosting guidelines cover configuration, privacy and monitoring, among other things, but updates, capacity management and resolving any incidents also remain the publisher’s responsibility. It is also important to ensure system maintenance doesn’t depend on a single expert.
With a technology partner like Relevant Digital, a publisher can use Prebid Server without maintaining its own server infrastructure. The partner handles the technical implementation and maintenance, leaving the publisher’s team free to focus on growing ad revenue.
When choosing a technology partner, it is important to assess how well the solution meets the publisher’s needs. Pay attention to, for example, the supported demand partners and ad formats, server locations, the scope of reporting and the technical support available. The solution's flexibility and the option to switch providers later are also important factors to consider.
For more on the differences between building your own solution and outsourcing, see our guide comparing building ad tech in-house with buying a ready-made solution.
Client-Side vs. Server-Side Prebid: How to Choose the Best Setup
The best header bidding setup for a publisher is found by comparing the options in practice. In a test run by Relevant Digital, we compared client-side, server-side and parallel bidding on both desktop and mobile. 25% of traffic was directed to client-side, 25% to server-side and 50% to parallel bidding. Several demand partners took part, including Adform, Equativ, Magnite, PubMatic and Xandr.
In the test, parallel bidding delivered the best ad revenue, but client-side came very close. Parallel bidding’s advantage was clearer on desktop. Server-side earned slightly less but delivered the fastest page loads. The differences between the setups were, however, much smaller than in our earlier tests.
The biggest differences were in the performance of individual demand partners. Some partners achieved a higher eCPM server-side, even though they may have bid less often. On desktop, some partners took part in server-side auctions less often, while on mobile some partners bid more actively. The results show that the same setup is not necessarily the best for every partner or every environment.
That is why it makes sense to test the options with comparable traffic groups and assess the results over a long enough period. In a hybrid setup, each demand partner can be run in the environment where it performs best. Parallel bidding, in turn, can be used selectively for partners whose participation in both client- and server-side auctions delivers a measurable uplift.
Read more about the test results in our article “Prebid Testing: How Server-Side, Client-Side, and Parallel Bidding Compare”.
Manage and Optimize Prebid Server with Relevant Yield
Relevant Yield lets publishers choose the header bidding setup that suits their needs and manage it flexibly. HB Manager supports both Prebid.js and Prebid Server at no extra cost. Through its UI, ad ops and yield teams can manage auction settings and test different setups without constant help from developers.
HB Analytics, in turn, gives visibility into how auctions run and how demand partners perform, supporting data-driven optimization. This helps publishers identify their best-performing demand partners, compare the impact of different setups and grow their ad revenue based on measured results. The goal is to find the setup that works best for the publisher and to grow ad revenue through continuous, data-driven optimization.
Talk to the Relevant Digital team about how implementing Prebid Server or optimizing your current setup could help grow your ad revenue.

