What to audit when core web vitals are green but page rankings still drop: a practical checklist

I remember the first time I saw Core Web Vitals all green and felt triumphant—only to watch some of my most important pages slide down the SERPs a few weeks later. If that’s happened to you, you’re not alone. Green CWV is a great start, but it’s not a magic bullet. I’ve compiled a practical, field-tested checklist of things I audit when Core Web Vitals look healthy but rankings still fall. Work through these items methodically and you’ll usually uncover the real problem.

Start with data — confirm the drop and its scope

Before anything else, validate what’s happening and where:

Check Google Search Console for changes in impressions, clicks, and average position for affected pages over at least 90 days.

Compare organic traffic in Google Analytics (or GA4) across the same time windows to rule out tracking issues or seasonality.

Use your rank tracker (Ahrefs, Semrush, SERPWatcher) to see which keywords and pages lost positions—are losses concentrated or broad?

Sometimes what looks like a ranking drop is actually a reporting artifact or a decline for a subset of queries. Confirm the problem before digging deeper.

Check for algorithm updates and SERP volatility

I always check whether a known Google broad core update, spam update, or product review update rolled out around the time of the drop. These updates can change how Google weights content, E-E-A-T, or intent matching.

Look at update trackers (Moz, Search Engine Journal, Semrush Sensor) and Google’s official Search Central announcements.

Inspect SERP features: did featured snippets, "People also ask", or local packs expand and push organic results down?

Audit content relevance and intent

Even with great CWV scores, content can be mismatched with user intent. When intent shifts (e.g., from informational to transactional), pages that previously ranked well can lose visibility.

Re-evaluate target keywords: are you still matching the primary intent? Use SERP analysis—look at top-ranking pages and see what format they use (listicle, product page, long-form guide).

Refresh content: add recent stats, examples, or unique insights. Thin or outdated content is a frequent culprit.

Check for cannibalization: multiple pages targeting the same query can hurt rankings—consolidate or canonicalize where appropriate.

Review on-page SEO and metadata

Meta titles, descriptions, H1s, and structured data still matter for CTR and relevancy signals.

Ensure titles and meta descriptions are aligned with the search intent and include primary keywords naturally.

Audit H1–H3 structure for clarity and avoid keyword stuffing.

Revalidate structured data (schema.org): errors can prevent rich results and reduce visibility.

Inspect technical SEO beyond Core Web Vitals

I’m careful to separate CWV from other technical factors. Pages can be fast and stable yet not crawlable or indexed correctly.

Indexation: Use Search Console Coverage and the “inspect URL” tool to confirm pages are indexed and there are no indexation warnings or soft-404s.

Crawlability: run a crawl with Screaming Frog or Sitebulb to detect blocked resources, noindex tags, or robots.txt rules affecting pages.

Canonical tags: ensure canonicalization points to the correct URL and there aren’t accidental self-canonical loops or cross-domain canonicals.

Redirect chains: eliminate long redirect chains and ensure 301s are used correctly.

Backlink and referral profile checks

Changes in your link profile or competitors gaining strong links can impact rankings quickly.

Use Ahrefs or Majestic to detect sudden lost backlinks or toxic links that might have appeared.

Compare competitor link acquisitions—if a rival gained high-authority links, they could have leapfrogged you.

Consider outreach to recover lost links or earn new, highly relevant backlinks.

Examine UX and conversion signals beyond CWV metrics

Search engines increasingly use engagement signals (dwell time, pogo-sticking) to judge satisfaction.

Analyze bounce rates, session duration, and pages per session for impacted pages.

Test mobile UX personally: navigation, intrusive interstitials, and distracting ads can harm perceived page quality.

Run user tests or session recordings via Hotjar or FullStory to spot friction that analytics alone won’t show.

Check for duplicate or syndicated content issues

Duplicate content—internal or syndicated—can dilute rankings.

Scan for duplicate content with Siteliner or Screaming Frog. Fix duplicates with canonical tags, 301 redirects, or by consolidating pages.

If you syndicate content, ensure canonical tags point to your original, or use noindex on syndicated copies.

Validate structured data and rich result eligibility

Schema errors can cause you to lose rich snippets, which often drive higher CTRs and perceived relevance.

Use Schema.org validators and Google’s Rich Results Test to check structured markup.

Fix warnings and malformed JSON-LD; even small parse errors can prevent rich features from showing.

Review server and hosting issues

Sometimes performance looks good in lab tests but the origin server has intermittent problems.

Check server logs for 5xx errors, spikes in latency, or timeouts.

Confirm CDN configuration and caching rules aren’t serving stale content or inconsistent headers.

Audit internal linking and site architecture

Internal links distribute authority and help Google understand page importance. A recent change here can shift rankings.

Ensure important pages are well-linked from category pages and the main navigation.

Look for orphaned pages or pages deeply buried under many clicks from the homepage.

Look at localization, hreflang, and canonical cross-country issues

If you operate in multiple languages or markets, misconfigured hreflang can lead to the wrong country or language page being shown.

Validate hreflang tags and make sure language and country code mappings are correct.

Check geo-targeting settings in Search Console for domain or subfolder targeting.

Confirm no manual actions or security issues

Manual penalties or hacked content will tank rankings regardless of CWV.

Check Search Console for manual action notices and security issues (malware warnings).

Run a security scan to ensure there are no hidden injected links or content.

Use server logs and search console data for deeper clues

Server logs reveal how Googlebot behaves on your site. I often find crawl budget changes or unexpected blocking that aren’t visible elsewhere.

Analyze crawl frequency for affected pages—has Googlebot reduced visits?

Check for spikes in crawl errors or long response times in logs around the drop period.

Audit Area	Quick Check	Priority
Intent & Content	Compare top SERP pages to yours	High
Indexation/Crawl	Search Console + Screaming Frog	High
Backlinks	Lost/gained link analysis	High
Structured Data	Rich Results Test	Medium
Server/Hosting	Log file & uptime checks	Medium
User Signals	Analytics & session recordings	Medium

Tools I rely on for these audits: Google Search Console, Google Analytics / GA4, Screaming Frog, Ahrefs, Semrush, PageSpeed Insights, Chrome DevTools, Hotjar/FullStory, and server log analyzers. I typically create a prioritized action plan from the checklist above and tackle the highest-impact items first—content intent fixes, indexation problems, and lost backlinks usually move the needle fastest.

If you want, send me the URL structure of the affected pages (or a few sample URLs) and I can point you to the most likely culprits based on the patterns I’ve seen. I often spot the issue within a few minutes of looking at the SERP and page characteristics.