Third parties per page
How many distinct third-party services the page embeds.
At a glance the headline numbers for Third parties per page
How many distinct third-party services the page embeds.
The typical site embeds 3 distinct third-party services.
Distribution & median CLS site count and median CLS at each level of third parties per page
Passing CLS by third parties per page which level passes the CLS most often
Third parties per page 3. p75 6. p99 14. At the low end (0): CLS 0.01. At the high end (>p98): CLS 0.05. computed
Why this matters for the Core Web Vitals, and where to start fixing it
Every third party on the page is code you invited but do not control. Each one adds connections, main-thread work, or both. The count is the simplest measure of how much of your page's fate is outsourced.
Counts only creep upwards. Tags get added for a campaign and survive it by years. Walk the list and remove every service nobody can name a reader for. Most sites can drop several without anyone noticing anything except better vitals.
How do third parties affect the Core Web Vitals?
Third parties correlate with the LCP. With a couple of services, 88% of sites pass the LCP. On the most tag-heavy sites, 83% do. Third parties have little causal relationship with the LCP. That makes having many third parties more likely a result or an indication of a slow, heavily built site than a cause of a slow one.
The effect is bigger on CLS. With a couple of services, 82% of sites pass it. On the most tag-heavy sites, 72% do.
Chrome field data from 94,910 sites, representing millions of real page loads. How we measured.