The Impact of Lazy Auctions on Header Bidding: A/B Testing Results

While looking for a way to increase your programmatic ads performance to benefit from them more, you might have heard of Lazy Auctions. As it is a relatively new technology in the world of digital advertising and is rapidly gaining popularity as companies look for ways to optimize their advertising campaigns and generate more revenue. We’ve conducted A/B tests to check its real abilities in terms of improving performance. 

Keep reading to find out the results. 

What is Lazy Auction?

A Lazy Auction is a feature in programmatic advertising that delays the auction process until the ad slot is likely to be rendered. It works by holding back ad requests until buyers have enough demand to fill the available inventory. By doing so, Lazy Auctions can increase ad inventory’s value by reducing unnecessary requests, which can lead to a more efficient and profitable advertising ecosystem. 

As we see potential in enhancing the performance of the ads and, therefore, the revenue, we’ve added a Lazy Auction feature to our header bidding tool Quick Wrap. It works based on the “auction per ad unit” feature. Both methods query only the ad slots available on a given pageview, not all of the ad slots listed. What defines Lazy Auction is splitting one auction into several, depending on the location of ad units on the page. 

Influence of Lazy Auctions on Header Bidding – benefits for publishers

Advertising networks often send overwhelming bid requests to SSP platforms, causing overloading and potential revenue loss from header bidding. To combat this issue, publishers are often asked to reduce the number of bids they send. However, using Lazy Auction can provide a more efficient solution by reducing the number of bids with little chance of displaying an ad and easing the burden on SSPs. 

Additionally, Lazy Auction increases the likelihood that bid responses from SSPs will enter the auction and not be wasted on ad units that are rarely displayed, such as those that require extensive scrolling. This makes Lazy Auction particularly effective on long vertical pages with many placements. 

Do Lazy Auctions help publishers increase revenue? – Yieldbird’s Research

The research was conducted in our header bidding tool QuickWrap, where we can run A/B tests. Our experiment involved dividing traffic into four separate samples: 

• Sample 1 – without Lazy Auctions 
• Samples 2,3, and 4 used Lazy Auctions, each with different parameters enabled to determine the optimal auction triggering method. 

We tested these parameters across three domains: a sports site with short pages and two long multi-scroll news pages. Using A/B testing in QuickWrap, we collected data to help users make informed decisions and improve advertising campaigns.

Lazy Auctions can be set in many ways, and as we may see on the chart below, only some settings (and we are testing three on each site here) work positively. The feature requires some optimization and testing to adjust the parameters of the page. Nevertheless, there was one good setting on each, generating more revenue from 1-4%, and these are the best versions we will focus on further. 

Further testing showed Lazy Auction can reduce the number of Ad Requests. Our research has shown that this is particularly evident in domains 2 and 3, where the best-performing v2 samples generated the most revenue (+3% and +4%), despite having fewer requests (-14% and -7%, respectively). This translates to a higher revenue per pageview on these samples. Although domain 1 did not significantly reduce requests, the best sample still had only 0.7% fewer requests than when Lazy Auctions were not utilized. 

Notably, the reduction in the number of requests resulting from Lazy Auction testing also affects the request-CPM metric, which would be overstated. For that reason, we ignore this metric in this particular test.

Our research has shown that Lazy Auction has a significant impact on increasing header bidding’s share of total revenue. The best-performing versions increased between 3% and 35%, as seen in the charts below.

The increase in header bidding participation can be attributed to the influence of Lazy Auctions on buyers’ bid wins and win rates in header bidding. This can be seen in fewer bids on Lazy Auction samples than those without them. Reducing unnecessary bid requests gives buyers a higher chance of the ad actually rendering therefore increasing the number of won impressions. This, in turn, leads to greater participation in header bidding and a boost in revenue.

In this case, fewer bids do not translate into fewer auctions won. On the contrary, the win rate and the wins’ absolute value are higher.

Do Lazy Auctions increase programmatic monetization?

After conducting our research and analyzing the data, we have concluded:

• The implementation and optimization of Lazy Auction can significantly increase revenue.
• Lazy Auction positively affects Header bidding, as it eliminates unnecessary bids, increases the chance of winning, and ultimately boosts HB’s share of the revenue. 
• By carefully launching and optimizing Lazy Auction, companies can improve their advertising campaigns and generate more revenue, making it a valuable tool in digital advertising.

As you see, there are plenty of pros to Lazy Auctions. Would you like to check how lazy auctions impact on header bidding by yourself? The A/B tests and implementation were done by the QuickWrap tool. Schedule a call with our experts and discover the possibilities of our tool and the rest of our programmatic solutions. 

LET’S GET IN TOUCH!