Core Web Vitals Green But Rankings Still Dropping

All-green Core Web Vitals and falling rankings is not a contradiction; it is a diagnosis. Passing CWV is a low-weight page-experience input that keeps you out of a hole, not a lever that climbs you out of one. When positions slide while every metric stays green, the cause is almost always content relevance, depth, or author authority, not speed. The tell that confirms it is where the decline concentrates. A loss spread across one section while the rest holds is a content-market-fit problem, not a technical one, and six more months of performance tuning will not touch it.

CWV is a tiebreaker, not a primary lever

Google’s own framing is that page experience, including Core Web Vitals, is one of many signals and is not the dominant one. Content relevance and quality decide who is eligible to rank; CWV can act as a tiebreaker among pages that are otherwise comparable. The verified thresholds remain LCP at or under 2.5 seconds, INP at or under 200 milliseconds (INP replaced FID in March 2024), and CLS at or under 0.1, all measured at the 75th percentile of field data.

The implication is mechanical. If your content is a clear step behind what now ranks, no amount of speed closes that gap, because the two pages were never tied. Spending two quarters shaving milliseconds off an already-green page is effort aimed at a signal that was never the bottleneck. The page was not losing on the tiebreaker; it was losing on the main event.

The decline-pattern diagnostic

Before touching anything, map where the loss lives. Segment the decline by section in Search Console. The pattern tells you the disease.

A uniform, site-wide drop points toward something systemic, often technical or a broad quality reassessment. A drop concentrated in one section while others hold steady points squarely at content. If your blog is collapsing while product or service pages are stable, you do not have a site-speed problem; you have a content-market-fit problem in the blog. The pages that held were never the issue, and the pages that fell share something the survivors do not.

This ten-minute segmentation is the step teams skip when they latch onto CWV. The metric was visible and fixable, so it became the assumed cause, when a quick look at the decline pattern would have pointed elsewhere.

The reason this misdiagnosis is so common is that CWV is legible. It produces a clear pass-or-fail with a color, a number, and a checklist of fixes, while a content-depth or authority gap is fuzzy and uncomfortable to confront. Teams gravitate to the problem that has a dashboard. The discipline is to resist treating the most measurable thing as the most important thing, and to let the decline pattern, not the convenience of a metric, decide where the work goes.

The competitive bar was raised under you

Content rarely declines in a vacuum. More often the bar moved. A page that earned its ranking years ago against the content that existed then is now being compared against everything published since. The SERP for that query has been re-contested, and the winners cover the topic more completely, more currently, and more credibly than your page did when it first ranked.

Run the test directly. Take your most-declined page and compare it head-to-head against the pages now ranking above it for its main query. You are looking for what they cover that you do not, how current they are, and who is credited as the author. The gap is almost always visible in minutes, and it is almost always depth, freshness, or authority rather than load time.

Be specific about what “raised bar” means in practice, because the phrase is easy to nod at and hard to act on. Open the top three results side by side with your declined page and inventory the subtopics each one covers. You are looking for the questions the winners answer that yours leaves untouched: a sub-question you skipped, a worked example you omitted, a recent development you never folded in, a data point that has since changed. Note the publish or last-updated date on each competing page; if they were refreshed in the last year and yours has not been touched since it first ranked, freshness alone may explain part of the slide. The output of this exercise is a concrete list of gaps, not a vague sense that the competition got better, and that list is what you act on rather than a hunch.

E-E-A-T and author authority on YMYL

For finance, health, legal, and other your-money-or-your-life topics, author authority is a heavier factor than elsewhere, and it is a common silent cause of decline. E-E-A-T (experience, expertise, authoritativeness, trustworthiness) is the current framing, and on YMYL content Google weights signals of who is behind the page.

A page bylined to a faceless company name, or to no one at all, competes poorly against one written by a credentialed, identifiable author with a real bio and a track record on the subject. If your declining YMYL pages carry a generic byline while the pages overtaking them carry named experts, the authority gap is doing the damage. This is not something CWV or word count fixes; it is an identity and credentials foundation that has to be built before depth even counts.

Prioritize: identity, then depth, then features

The fix order matters because the foundation gates everything above it. For YMYL especially, establish author identity and credentials first; depth published under an anonymous byline still struggles to earn trust on sensitive topics. With identity in place, rebuild depth on the pages that lost the most, matching or exceeding what the current winners cover rather than chasing an arbitrary length target. Interactive features and finer enhancements come last, after the substance is competitive.

Triage by business value, not by ease. Fix-everything is a misallocation; the page that lost the most traffic and matters most to revenue earns the first renovation, not the page that happens to be quickest to edit. Word-count contrasts are illustrative only. A thin page often loses to a more complete one, but the goal is genuine coverage of what the query needs, not hitting a number.

The misallocation framing matters because the CWV spend was not wasted in a vacuum, it was spent on the wrong lever at the wrong time. Green Core Web Vitals are still worth having; the error was sequencing performance work ahead of the diagnosis that would have revealed a content and authority gap. Reframing it as misallocation rather than waste keeps the team honest about the real lesson: diagnose where the decline concentrates before committing a quarter of engineering time to the metric that happened to be visible.

What to do

Stop treating CWV as the lever once it is green. Map which sections and pages declined in Search Console, then run head-to-head SERP comparisons on the worst losers to find the depth, freshness, or authority gap. For YMYL pages, fix author identity and credentials first. Then reinvest in genuine depth where the business value justifies it, prioritized by traffic lost rather than by what is easiest to ship. Recovery follows the content improving relative to the competition, not a date on a calendar.

Frequently Asked Questions

If CWV does not move rankings much, should I ignore it?

No. CWV is worth getting into the green because it removes a possible drag and acts as a tiebreaker among comparable pages. The point is sequencing: once green, additional speed work yields little, and the next gains come from content and authority. Treat it as hygiene you maintain, not the lever you pull to recover lost positions.

How do I tell a content decline from a technical one?

Look at the shape of the loss. A drop concentrated in one section while others hold is a content signal; a uniform site-wide drop suggests something systemic or technical. Pair that with a head-to-head SERP comparison of the worst-losing pages to confirm whether the issue is depth, freshness, and authority rather than crawling or rendering.

Sources

Understanding Core Web Vitals and Google Search results, Google Search Central: https://developers.google.com/search/docs/appearance/core-web-vitals
Web Vitals, web.dev: https://web.dev/articles/vitals