URLs that failed all three tests
Posted: Sat Dec 28, 2024 3:55 am
It’s also interesting to contrast with a performance metric which was not part of the Page Experience update: Speed Index, as reported in Lighthouse lab results. As “passing” the three thresholds for CWV represents being in the top 36.3% of URLs by that metric, we can compare what ranking difference is associated with being in the top 36.3% for Speed Index. Average rank of URLs with CrUX data - August 2021 showing core web vitals and speed index.
We can see in this chart that Speed Index, despite not being an explicit ranking afghanistan phone number database factor, has a modest improvement in average rank associated with this percentile breakdown (0.17, vs. 0.39 for passing all three CWV thresholds). This doesn’t mean that Speed Index is a ranking factor, it just means that these things can be related in more complex ways. (If you’re a mathematics nerd like me and you’ve just noticed the weighted average rank of the two groups is not the same, that’s because there are a tiny number of URLs for which I was able to obtain CrUX data, but not lab data, due to server errors, etc.
) So did anything happen? Actually, yes. But it’s more subtle. The real impact was felt for . Although these URLs often started out ranking best of all (probably because they disproportionately represent some important, household name brands) they’ve taken a hit with the update. These URLs have had a 1.15 position ranking drop, compared to around 0.2 for URLs with CrUX data taken as a whole. URLs passing one, two or three thresholds alongside their average rank on a graph.
We can see in this chart that Speed Index, despite not being an explicit ranking afghanistan phone number database factor, has a modest improvement in average rank associated with this percentile breakdown (0.17, vs. 0.39 for passing all three CWV thresholds). This doesn’t mean that Speed Index is a ranking factor, it just means that these things can be related in more complex ways. (If you’re a mathematics nerd like me and you’ve just noticed the weighted average rank of the two groups is not the same, that’s because there are a tiny number of URLs for which I was able to obtain CrUX data, but not lab data, due to server errors, etc.
) So did anything happen? Actually, yes. But it’s more subtle. The real impact was felt for . Although these URLs often started out ranking best of all (probably because they disproportionately represent some important, household name brands) they’ve taken a hit with the update. These URLs have had a 1.15 position ranking drop, compared to around 0.2 for URLs with CrUX data taken as a whole. URLs passing one, two or three thresholds alongside their average rank on a graph.