Core Web Vitals get all the attention because they're easy to measure and easy to obsess over. A dashboard turns red, someone panics, and a week disappears into shaving milliseconds off a metric that was never the real constraint.
In sixteen years of auditing sites, the thing actually holding a page back is rarely the one with the loudest score. It's usually further upstream — in how the page is structured, linked, and understood.
Where I actually look first
Before I touch performance, I check whether the page can even be found, crawled, and understood: internal links, canonicals, duplicate templates, and whether search engines and AI models can extract the point of the page at all.
"A perfect performance score on a page nobody can find is a fast route to nowhere."
Once the fundamentals are solid, performance work compounds. Do it first, and you're often optimizing a page that was never going to rank regardless of how quickly it loaded.
A quick checklist
- Is the page internally linked from somewhere that matters?
- Does the canonical point where you think it does?
- Can an AI model actually extract what this page is about?
→ I run this exact triage with my own crawler — full-site signals plus an AI-readiness grade — so I can see all of it on one screen before I recommend a single change.
None of this means Core Web Vitals don't matter. It means they're the last mile, not the map. Fix the foundation, then make it fast.
