UX Rating answers one question: how much of the experience your site delivers is good? It is the percentage of your real-user measurements that meet Google’s good thresholds. The higher the score, the more of your real-world traffic is fast, stable and responsive.
How is UX Rating calculated?#
Each real user session reports up to four measurements: Largest Contentful Paint, Cumulative Layout Shift and Interaction to Next Paint (the three Core Web Vitals), plus Time to First Byte. Each measurement is rated good, needs improvement or poor against Google’s thresholds.
The system pools all individual metric ratings across all sessions and calculates what percentage fall into each bucket:
- Good UX — the percentage of all metric ratings classified as “good”
- To Improve UX — the percentage classified as “needs improvement”
- Poor UX — the percentage classified as “poor”
UX Rating is not a per-session pass/fail. It’s an aggregate across all four metrics and all sessions, so a session with one slow interaction contributes one poor rating out of four rather than a failed visit. For example, if you had 1,000 sessions and each reported 4 metrics, that’s 4,000 individual ratings. If 3,200 of those were "good", the Good UX percentage would be 80%.
Missing metrics are excluded, not penalised#
If a visitor loads a page but never interacts with it (no click, tap, or keypress), there’s no INP value for that session. That session still contributes its LCP, CLS, and TTFB ratings — but doesn’t drag INP down with a zero or a penalty. UX Rating is weighted toward metrics that are actually observed for each visit.
Where to find UX Rating#
UX Rating comes from field data, so it shows up throughout Calibre wherever CrUX or Real User Monitoring data appears:
- Across all your Sites: the Sites list shows UX Rating for every Site at a glance.
- Real User Monitoring: the Dashboard, Pages and Audience reports.
- CrUX: the Dashboard, Pages and Audience reports.
The Audience report gives the clearest view: UX Rating as overall percentages of Good, Needs Improvement or Poor, with a bar chart over time. CrUX shows the same split as its experience distribution.
How to improve your UX Rating#
Since UX Rating is a composite of four metrics, improving any one of them will lift your overall rating.
Focus on the metrics where the largest share of ratings fall outside "good":
- Improve Largest Contentful Paint — optimise images, reduce server response times, eliminate render-blocking resources.
- Improve Cumulative Layout Shift — set explicit dimensions on images and embeds, avoid inserting content above existing content.
- Improve Interaction to Next Paint — reduce JavaScript execution time, break up long tasks, optimise event handlers.
- Improve Time to First Byte — use a CDN, optimise server-side processing, enable caching.
Use segmenting and filtering to locate Pages, Elements, or Page Groupings that have:
- The highest number of views (affecting more users)
- Poorly rated performance (indicated in red)
Using this approach you can locate the largest impact issues, and fix them, one at a time. By fixing the highest traffic, most glaring performance issues you will quickly improve poor experiences to good experiences. Your visitors will notice immediately.
Frequently asked questions#
Why Core Web Vitals and Time to First Byte?#
Core Web Vitals are the current gold standard set of performance metrics in the browser. They strongly correlate with user perception of both the speed and overall quality of a users experience.
Time to First Byte is a critical metric because it highlights a delay that users feel on every visit to your site. A fast TTFB keeps users engaged and means pages can initialise more quickly.
Combining these metrics using aggregation we’re able to quickly understand overall render speed, page stability and interactions.
Google’s Core Web Vitals assessment also partially overlaps with UX Rating in that your pages are evaluated for percentage of Good user experiences.