Third Party Web: an analysis of third-party script costs
Third Party Web is an initiative to analyze the cost of third-party scripts on websites to provide Internet users and developers with actionable information.
The project has four simple goals:
- Quantify the impact of third-party scripts that run on the top 1 million sites according to Alexa.
- Identify scripts that have the greatest performance costs.
- Provide developers with information.
- Incentivize responsible script behavior.
The project team runs two scans on about four million sites per month using Lighthouse on mobile to identify third-party scripts and the performance impact they have.
According to the data, about 65% of all script execution time is caused by roughly 800 origins; the top 100 origins account for about 59% of all script execution time on the analyzed sites.
Read also:Â Extensions may impact performance.
The project sorts scripts into categories such as ads, social, analytics, or video to make it easier to compare the performance impact of related scripts.
The biggest offenders in regards to the average performance impact:
- CreateJS CDN -- Libraries -- 3188ms on average
- WordAds -- Advertising -- 2543ms on average
- Popads -- Advertising -- 1245ms on average.
- 33 Across -- Advertising -- 1170ms on average.
- Wix -- Hosting Platforms -- 1153ms on average
All scripts, with the exception for the Wix script, which was found on more than 158k sites, were found on a relatively low number of sites (about 30,000 or less).
The biggest offenders in regards to distribution:
- Google/Doubleclick Ads -- Advertising -- 1412404 executions, 330ms average.
- Google Tag Manager -- Other -- 1093167 executions, 386ms on average.
- Wix -- Hosting Platforms -- 158466 executions, 1153ms on average.
- Facebook -- Social -- 1212567 executions, 120ms on average.
- Google CDN -- Libraries -- 811231 executions, 178ms on average.
Advertisement and Mixed/Other scripts make up the largest chunk of third-party script executions.
It should not come as a surprise that scripts by major Internet companies -- Google and Facebook specifically -- are found on a large portion of scanned sites. Google alone has five scripts in the total impact top ten, with three of them breaking the one million execution barrier.
Google scripts were found in about one in three sites at a minimum, Facebook scripts in one in four sites.
Closing Words
Third-party scripts impact web performance significantly; a simple visual comparison of the loading performance of sites with and without content blockers is enough to highlight that fact.
The result of the study is not really that surprising: third-party scripts impact performance, and Google and Facebook have scripts running on a large portion of Internet sites.
Now You: What is your take on the study?
http://www.ghacks,net
has 4 different google trackers
I don’t block them!
For years, I’ve blocked all Google, Facebook, Twitter, Amazon, and LinkedIn third-party calls.
Everything worthwhile works great, and it’s much faster and significantly more private.
I control ghacks’s third-party shit by setting ghacks’s CSP to this policy — default-src ‘self’ ‘unsafe-inline’ ‘unsafe-eval’ so no third-party crap at all.
I tried browsing this ghacks.net page with a new Firefox profile and found third-party connections to more than eighty (80) different domains. Unbelievable… Then simply installing ublock origin in its default configuration makes this count drop to zero and yet everything works fine.
Ofcourse, allow third-party content is opening pandora’s box, you’re fucked once you do that.
“What is your take on the study?”
Forgotten soon like this one: https://github.com/snyderp/web-api-manager
As a user, just use a content blocker.
That’s all.
Yes but with multiple ad and tracking blockers, what are the left-over scripts?
“What is your take on the study?”
It seems plausible to me.
Personally, I don’t care about script performance much. I block scripts because they can’t be trusted. I have to assume that any third-party script is spying on me.
This helps reinforce my thoughts that most of the worth of blocking these outside scripts is worth more than whatever they are trying to present. Its gotten kind of silly how much work web hosters (wont even call them designers at this point) go just to try and profit by showing other people’s code/script/content.
There are a few places that hand pick their advertisers and forego 3rd party advertising altogether. Those I whitelist, because at least they are attempting to keep with the theme and have a hope of presenting me with something Im actually interested in.
For all the hype about their networks and how much they ‘know’ .. not even FB within their own site ever shows me ads for something I would be interested in and didn’t already know about. All they do is show me variants of things Ive already found, which is not useful, and certainly isn’t as all knowing as they are being promoted.
Google penalizes sites for poor performance. Except for the most popular ones (Facebook etc). Too much $ involved.
How do you spell hypocrites?
“Using site speed in web search ranking
Friday, April 09, 2010
You may have heard that here at Google we’re obsessed with speed, in our products and on the web. As part of that effort, today we’re including a new signal in our search ranking algorithms: site speed. Site speed reflects how quickly a website responds to web requests.
Like us, our users place a lot of value in speed — that’s why we’ve decided to take site speed into account in our search rankings.”
@Ken Saunders:
“You may have heard that here at Google we’re obsessed with speed, in our products and on the web.”
I believe that *was* in fact true back in 2010. If you’ve used any of the “new” versions of their Web services — Gmail, Google Maps, Google Contacts, Google Tasks, etc. — you’ll realize to a certainty that that’s no longer the case.
Also, I’m going to go ahead and guess that Google somehow excluded the time it takes Google scripts to run from the speed-based component of their result rankings…
Thumbs up
I’d suggest all uBlock users to learn and use its “medium mode”, instead of default settings + extra filter lists. It does take some time to setup and familiarise with, but the results will be quite rewarding.
https://github.com/gorhill/uBlock/wiki/Blocking-mode:-medium-mode
gotta luv goygle, really