Does Core Web Vitals Actually Move Rankings? A 240-Domain Study.

The question

Google has repeatedly described Core Web Vitals (LCP, CLS, INP) as a ranking factor — but as a "tiebreaker" with limited weight. Marketing teams repeatedly ask us: is it worth the engineering cost to fix?

We ran a controlled analysis on 240 client domains across 24 months. On 42 of those domains, we completed meaningful CWV remediation during the window. On 198, we didn't change CWV materially. This gave us a rough treatment/control structure.

Headline findings

What drove the lift

Breaking down the 42 remediated domains:

• Domains with failing LCP (>2.5s): Median lift of 14% after remediation.
• Domains with CLS issues (>0.1): Median lift of 7% after remediation.
• Domains with INP issues (>200ms): Median lift of 6% after remediation.
• Domains with multiple failing metrics: Median lift of 22% after remediation.

The lift wasn't primarily from Google's ranking signal. We looked at individual keyword position changes and found position changes were real but small (median +1.2 positions on previously ranking keywords). The majority of the traffic lift came from improved user behavior — higher click-through from SERPs when pages render fast, lower bounce rates, higher pages-per-session, and better conversion.

What didn't work

We also tracked attempts that failed. The most common patterns:

• Chasing lab scores instead of field data. PageSpeed Insights lab scores can move dramatically without changing CrUX field data. Only field data moves rankings.
• Fixing one metric while breaking another. We saw multiple cases where LCP improvements caused CLS regressions.
• Fixing non-indexed templates. Several teams invested heavily in improving CWV on templates with minimal organic traffic. The ROI was near zero.
• Over-engineering past the threshold. Moving LCP from 2.4s to 1.1s on already-passing templates did not produce measurable ranking lift.

Recommendations

• Measure field data, not lab scores. CrUX report, GSC, or RUM — not synthetic PageSpeed Insights runs.
• Prioritize failing templates with high organic traffic. Don't optimize what doesn't rank.
• Fix all three metrics together. Partial remediation often causes regressions elsewhere.
• Don't over-invest past passing thresholds. After you're green, additional CWV work has diminishing returns. Invest resources in content or UX instead.

Methodology note: Treatment and control groups weren't fully randomized — remediation was prioritized based on client need, so there's some selection bias. We controlled for industry, baseline traffic, and content investment, but the findings should be interpreted directionally, not precisely.