If I want to see the impact of a third party script/tag has on page load times should I use the SPOF tab or Block tab?
What is the difference between SPOF and Block?
How come when I run a test in blocking a tag, the page load times go up and not down which you would expect? Any ideas?
Block is a fast-fail while SPOF routes the requests to a black hole to timeout.
As for why times go up, if it is in Chrome it is likely because of how the blocking is done currently. It uses an extension that runs like an adblocker and Chrome has a crazy-amount of overhead on the extension API.
There is a script command “blockDomains” that works at the DNS level instead and doesn’t add the overhead but it only works on domains, not URL substrings. I’m considering adding a UI field to expose the domain-level blocking and reduce the usage of the existing block.
thanks for the reply. The “blockDomains” sounds interesting. Any advice on when to use SPOF versus Block?
I usually use block when I’m trying to evaluate the impact of a given component to performance. Like “how fast is it without the ads” or “without the a/b testing” to judge the performance impact. I use SPOF to see “what is the impact to performance if the ads provider is unavailable” or to test all of the 3rd-parties to make sure there are no hard dependencies.