AI-generated Key Takeaways
-
This audit focuses on identifying and mitigating long-running requests that delay the initial ad request, ultimately hindering ad loading speed.
-
By reducing or removing these "blocking requests," publishers can significantly improve the latency of their first ad request.
-
Optimization strategies include eliminating unnecessary requests, deferring non-critical requests, parallelizing request execution, and enhancing backend service response times.
-
The audit highlights the top 5 most impactful blocking requests based on a calculated "cost" that factors in request duration and self-time.
-
Requests with self-times exceeding 250ms or durations longer than 1s are flagged for potential optimization to improve ad loading performance.
Overview
This audit identifies long-running requests that are blocking the first ad request. These are requests that were initiated and which also received a response before the first ad request was made. Reducing the number and duration of these blocking requests will reduce the latency of the first ad request, speeding up ad loading.
Recommendations
The details of this audit contain the top 5 blocking requests (by cost). The goal here is to remove or reduce the duration of these requests to speed up ad loading. Some tips for doing this include:
- Eliminating unnecessary requests.
- Deferring non-critical requests until after ads are loaded.
- Issuing requests in parallel rather than serially.
- Improving response times by optimizing backend services, using HTTP/2, etc.
More information
This audit displays the top 5 bottleneck requests by cost. The cost of a request is computed as follows:
request duration + (request self-time * 3)
Where request self-time is equal to duration minus time spent blocked by other requests. Only requests with a self-time greater than 250ms or a duration greater than 1s are considered.