Đoàn Lê Minh Anh
Website performance optimization is about more than making a page load quickly on your development team’s computers. What matters is understanding how real users experience the site across different devices, networks, and usage conditions. Chrome UX Report CrUX gives businesses insight into those real-world experiences, rather than relying only on simulated test results.
This article explains how to read CrUX data, understand Core Web Vitals, and turn the findings into a practical optimization plan. By identifying which metrics are underperforming, which groups of pages are affected, and which changes could have the greatest impact, businesses can focus their website improvement efforts more effectively.
What Is Chrome UX Report CrUX?
Chrome UX Report, commonly known as CrUX, is a public dataset that aggregates real user experience metrics for eligible websites. Data is collected from Chrome users who have enabled the option to share usage statistics and crash reports, then aggregated in a way that protects privacy. Because it reflects real visits, CrUX offers a different perspective from tools that run tests in controlled environments.
CrUX does not measure every visit from every browser. Its coverage depends on eligible Chrome users and the amount of data available to aggregate. As a result, a page or group of pages with low traffic may not appear fully in the report. Data is also typically viewed over a rolling 28-day window, so recent changes may not show up immediately.
It is important to distinguish CrUX data from lab data. A tool such as Lighthouse runs a one-time test with a specified device and network configuration, while CrUX aggregates results from real-world experiences. The two sources do not necessarily match: a page may score well in a simulated test yet still cause problems for people using older phones or slower connections.
Key Metrics for Evaluating Core Web Vitals
Core Web Vitals are a set of metrics focused on important aspects of the user experience. When reviewing a CrUX report, pay attention to the three metrics below and consider the share of users with a good experience rather than looking at just one average value.
- LCP (Largest Contentful Paint): measures how long it takes for the largest content element in the viewport to finish loading. A good result is 2.5 seconds or less.
- INP (Interaction to Next Paint): evaluates how quickly a page responds to interactions such as clicks, taps, or typing. A good result is 200 milliseconds or less. INP replaced FID as a Core Web Vitals metric in March 2024.
- CLS (Cumulative Layout Shift): measures unexpected layout shifts while a page loads and is being used. A good result is 0.1 or less.
These thresholds are evaluated at the 75th percentile (p75), typically reported separately for mobile and desktop. Put simply, to achieve a good rating, at least 75% of visits need to fall within the relevant threshold. For example, an LCP p75 of 2.3 seconds means that most visits meet the good threshold, although some users may still experience slower loading.
Do not judge a website as good or poor based on a single aggregate score. Check each metric, device type, and page scope. One weak metric can undermine the experience even if the other two meet their thresholds. In addition to Core Web Vitals, the report may provide other details, such as connection type or load-time distribution, that add context to your analysis.
How to View and Interpret CrUX Data
1. Start with PageSpeed Insights
Enter the URL you want to check in PageSpeed Insights to view field data if the URL or website has enough CrUX data. The field data section shows whether the URL or origin is rated good, needs improvement, or poor for each metric. Lab tests appear separately below; they can help uncover technical clues, but they do not replace field results.
If there is no data for a specific URL, check the report at the origin level, which covers the entire domain. Keep in mind that origin-level data combines experiences across multiple pages, so you cannot assume every page performs the same way. Use the report to understand overall trends, then check important page templates such as product pages, articles, category pages, and conversion forms.
2. Use the CrUX Dashboard to Track Trends
The CrUX Dashboard visualizes data over time and is often useful for monitoring a website at the origin level. Its charts can reveal whether metrics are improving or declining across different periods. Because the data is based on a 28-day window, the most recent dates may not reflect a change that was just implemented. Avoid comparing periods that are too close together and drawing conclusions too quickly.
3. Cross-Check Search Console and Internal Data
The Core Web Vitals report in Google Search Console groups URLs with similar statuses, helping teams identify clusters of pages that need attention. Combine it with traffic analytics, conversion rates, and error data to understand the business impact. For example, an issue affecting checkout pages may deserve higher priority than the same issue on a page with little traffic.
For a systematic analysis, record the report scope (URL or origin), device, time period, and p75 for each metric. Keeping track of this information makes before-and-after comparisons more accurate and helps prevent confusion between field data and simulated test results.
Using CrUX Data to Improve Core Web Vitals
CrUX data shows which issues users are experiencing, but it generally does not identify the specific technical cause. Use the report as a signal to form a hypothesis, then combine diagnostic tools and testing to confirm the cause. A practical process could include these steps:
- Identify the affected page groups and users. Review metrics by device and URL group; prioritize pages with high traffic or a strong connection to important business goals.
- Pinpoint likely causes. For LCP, check large images, server response time, render-blocking resources, and font loading. For INP, investigate long JavaScript tasks, third-party code, and interaction event handling. For CLS, review image dimensions, ads, dynamically inserted content, and fonts.
- Choose measurable changes. For example, optimize images for the right dimensions and format, defer unnecessary code, break up JavaScript tasks, or reserve space for components that load later.
- Test before release. Use lab data to catch problems quickly, and test on devices and networks that are close to your customers’ conditions.
- Check field data again. After implementation, allow enough time for the 28-day window to reflect the change, then compare p75 and the share of URLs rated good.
When advising businesses and implementing websites, LilyTech often recommends treating performance as part of digital product quality—not a one-time task. A good plan connects technical metrics with user behavior, conversion goals, and the team’s ability to maintain the improvements. If you make changes only to achieve a high score in a single test, the real-world benefits may not last.
Common Mistakes When Reading CrUX
- Assuming field data and lab data should match: they reflect different contexts, so use them together rather than comparing them mechanically.
- Evaluating a website using only one URL: one good URL does not represent every page template; conversely, origin data does not pinpoint exactly which page has a problem.
- Ignoring the 75th percentile: an average can hide users who are having a poor experience. Review the distribution and p75 for each device.
- Expecting results immediately after a fix: the 28-day data window means it takes time for the impact to become clear.
- Chasing thresholds at the expense of the experience: improvements to metrics should go hand in hand with keeping pages easy to use, content accessible, and conversion journeys smooth.
Conclusion
Chrome UX Report CrUX helps businesses understand website performance through real user experiences and provides a basis for prioritizing Core Web Vitals improvements. Check the data scope, analyze LCP, INP, and CLS by device and page group, then use technical testing to confirm the cause. Most importantly, track trends long enough to determine whether your changes improve the real-world experience.
If your business needs a website assessment, a performance improvement roadmap, or help connecting technical optimization with marketing goals, contact LilyTech for advice tailored to your current situation and team resources.
