1. Timeout Set Too Low
The timeout determines how long Prebid waits for bidders to respond before closing the auction and proceeding with the ad server request.
A timeout below 1,000 ms can be considered aggressive, as this type of configuration may exclude otherwise valid bid responses.
Values between 1,000 and 1,500 ms should also be reviewed. Although they do not necessarily indicate an error or misconfiguration, they may still be too short in certain regions, such as LATAM.
When using PBS—or Prebid Server—it is also important to verify that the server-side timeout (s2sConfig.timeout) is between 50% and 75% lower than the client-side auction timeout.
For example, setting the overall Prebid timeout and the Prebid Server timeout to the same value may cause server-side responses to arrive after the client-side auction has already ended.
TIP: The Timeout RTD Provider module can dynamically adjust the timeout based on factors such as the device, network connection, or auction complexity. Not using it is not necessarily an error, but it may represent an optimization opportunity.
2. User Sync Partially Disabled
User Sync allows bidders to recognize the same user context in future auctions, enabling them to measure and value each impression more accurately.
When User Sync is not working correctly, a bidder may experience a lower match rate, recognize fewer user contexts and, as a result, submit fewer bids or bids of lower value.
Many publishers use restrictive lists to define which bidders are allowed to perform User Sync. The problem arises when they add new bidders but forget to update the configuration.
Suppose a publisher has three bidders connected: Bidder A, Bidder B, and Bidder C.
However, its User Sync configuration is as follows:
userSync: {
filterSettings: {
iframe: {
bidders: ['Bidder A'],
filter: 'include'
}
}
}In this case, only Bidder A will be allowed to perform User Sync. Bidders B and C will be blocked, which will likely affect their performance.
As a general rule, unless specific legal or privacy restrictions apply, all bidders participating in the auction should be given the opportunity to perform User Sync.
3. High Ad Density
High ad density occurs when a page contains too many ad units or displays an excessive number of ads on screen at the same time.
This issue is often found in implementations where new placements are added to increase impression volume, without evaluating their cumulative impact on the user experience and the performance of each individual ad.
As a reference, an average of between 1 and 2.2 ads simultaneously in view may be considered reasonable. When this level is consistently exceeded, ads begin to compete with one another for the user’s attention.
Even when impressions maintain strong viewability levels, the available attention is divided among a greater number of creatives. This can lead to lower CTRs and reduce the individual value of each impression.
High ad density can also generate an excessive number of impressions per user within a very short period, particularly when combined with sticky formats, floating video ads, or refresh mechanisms.
Jounce Media uses ad impressions per user per minute as an indicator of ad load. It considers 5 to 8 impressions per user per minute to be within a normal or optimal range, while 10 or more impressions per minute represent a high-ad-load environment, typically associated with lower attention and weaker advertising performance.
For example, a page with several visible placements and automatic refresh can easily exceed eight impressions per user per minute, even when the user spends only a short amount of time on the page.
In these cases, an increase in impression volume does not necessarily produce a proportional increase in revenue. It may ultimately harm metrics such as CTR, attention, bidder performance, and revenue per session.
See you next time 🙌
