Lighthouse runs on a clean machine with a fast connection. CrUX measures your actual customers on mid-range Androids. We diagnose from field data, find the one or two elements costing you the pass, and fix those instead of chasing a score.
Three patterns account for most of the Core Web Vitals work we take on. If one of them is yours, the audit will confirm it in a fortnight.
We pass in Lighthouse but fail in Search Console.
Lab and field data measure different things. Search Console uses real visitors, which is the only version Google ranks on. We diagnose from the same source.
Field-first diagnosisWe optimised images and it barely moved.
Usually the LCP element isn't the image — it's a web font, a hero loaded by JavaScript, or a consent banner painting last. We identify the actual element per template.
LCP element traceThe page jumps while it loads.
Ad slots, late-loading banners and unsized images each contribute. CLS is fixable in an afternoon once you know which element is shifting and by how much.
Shift attributionSequenced deliberately — each phase depends on the one before it. Numbering here means order, not decoration.
Benchmarks, not promises. Your starting point decides how far and how fast — the audit gives you both numbers before you commit to anything.
The threshold Google uses. We target the 75th percentile, not the average.
Interaction responsiveness, measured on real devices rather than desktop.
Layout stability, checked per template group instead of sitewide.
Benchmarks reflect the standards we work to, not guaranteed outcomes.
They're a real but small direct signal — they rarely outrank relevance. The bigger effect is commercial: slow pages lose conversions before they lose rankings. We treat CWV as a revenue fix that happens to help SEO, which is the honest framing.
Lab tests vary with network conditions and machine load. Field data doesn't, because it aggregates thousands of real sessions over 28 days. If you're making decisions off a fluctuating Lighthouse number, you're chasing noise.
Some of it — image handling, script loading and CMS-level settings are often configurable. Deeper work like critical CSS extraction or hydration changes needs a developer. We'll be clear upfront about which bucket each fix falls into.
The report uses a rolling 28-day window, so expect roughly a month after the fix ships before the status flips — and it moves gradually as old sessions age out. Nothing is wrong if it doesn't change in week one.
Core Web Vitals rarely runs alone. These are the three we most often run alongside it.
Crawlability, indexation, site architecture and the render path Google actually sees.
Explore →GrowthCategory architecture, product templates, faceted navigation and the schema that sells.
Explore →FoundationScreaming Frog, Search Console and Semrush read together, ending in a ranked fix list.
Explore →No obligation, no sales sequence. We'll show you what's actually holding your Core Web Vitals back and what it would take to fix it.