Vitals Health Score

Vitals Health Score is an internal scoring model that compresses LCP, CLS, and INP into one headline number. Useful for dashboards. Useful for prioritization. Not a metric Google reports or ranks on directly.

That distinction matters. Google evaluates individual Core Web Vitals thresholds, not your blended score. If your team tells stakeholders “we improved Vitals Health Score from 68 to 82,” that can be fine for reporting. It becomes sloppy when people assume the score itself maps to rankings.

What it usually includes

Most teams build Vitals Health Score from the three Core Web Vitals: Largest Contentful Paint (LCP), Cumulative Layout Shift (CLS), and Interaction to Next Paint (INP). A common setup weights each metric, converts raw values into a normalized 0-100 scale, then averages them by template or URL group.

Tools vary. You might pull field data from Google Search Console (GSC) and CrUX, lab data from Screaming Frog with PageSpeed Insights integration, and debugging data from Lighthouse or Chrome DevTools. Some teams layer this into Looker Studio, while others keep it inside Ahrefs or Semrush reporting alongside rankings and traffic.

Why SEOs use it

Because executives do not want three percentile distributions and a lecture on p75 thresholds. They want one number, trend lines, and a clear answer on whether the site is getting better or worse.

For SEO teams, the score helps with triage. If category pages average 54 and editorial pages average 81, you know where to push engineering time first. If one release drops the score by 12 points on mobile templates, you have a regression signal fast.

It also helps tie technical work to business pages. Pair the score with revenue, organic sessions, or non-brand landing pages and you get a more defensible backlog than “site speed feels important.”

Where it breaks down

This is the caveat people skip: a single score can hide the actual problem. A page can have an acceptable blended score while still failing INP badly on mobile. Or it can pass lab tests and still fail field data in GSC because real users on weak devices get a different experience.

Another issue: scoring models are inconsistent. One company may weight LCP at 50%, another at 33%. So a score of 78 on one dashboard is not comparable to 78 somewhere else. That makes benchmarking messy.

Google's John Mueller has repeatedly pushed teams to focus on the underlying user experience rather than invented aggregate metrics. That's the right call. Use Vitals Health Score to manage work, not to pretend you found a hidden ranking KPI.

How to use it well

• Track it by template, device, and country. Sitewide averages hide failures.
• Keep the raw CWV metrics visible next to the score.
• Use field data first, lab data second. GSC and CrUX beat synthetic tests for prioritization.
• Set practical thresholds, like investigating any template below 70 or any 10-point drop after release.
• Correlate with outcomes that matter: organic entrances, conversion rate, and revenue per session.

If you want a clean internal KPI, Vitals Health Score is fine. Just do not confuse a reporting shortcut with how Google actually evaluates page experience.