Page experience update

A 79-day rollout, deliberately

Google announced the update months in advance, delayed it, then rolled it out gradually across nearly three months. The slow deployment was intentional and Google said so: the intention was to avoid sudden disruption.

For diagnosis that creates an unusual situation. There is no break point to find. A site affected by this did not drop on a date; it drifted over a quarter, which looks nothing like the step change the drop analyzer is built to detect.

What it actually weighted

Largest Contentful Paint, First Input Delay, Cumulative Layout Shift, mobile friendliness, HTTPS and the absence of intrusive interstitials. Google described it as a tie-breaker between pages of similar relevance rather than a factor that overrides content quality.

That description has held up. In audits, poor Core Web Vitals almost never explain a traffic collapse on their own. What they do is cap how well content improvements perform, which makes them worth fixing during a recovery but never the first thing to fix.

When Core Web Vitals are worth prioritising

• Never as the response to a sudden drop. A step change on a date is a ranking or policy event, not a speed problem that existed the week before too.
• Yes when you are competitively tied. Between two pages Google rates as similar, experience signals decide, and that is where they earn their keep.
• Yes when they are genuinely bad. Layout shift that moves a buy button is a conversion problem regardless of rankings.
• Note that FID was replaced by INP in March 2024, so audits written before then measure something Google no longer uses.

Source: Google ranking updates list