What Is Real User Monitoring?
Real User Monitoring (RUM) is a performance monitoring approach that captures and analyzes every transaction made by real users on a website or application. Unlike synthetic monitoring, which simulates user interactions from controlled environments, RUM collects data from actual browser sessions, providing authentic insight into user experience across diverse devices, networks, and geographies.
RUM works by injecting a small JavaScript SNI into your pages. This snippet records navigation timing, resource loading, JavaScript errors, and user interactions, then sends the data to a collection endpoint for aggregation and analysis.
RUM vs Synthetic Monitoring
Both approaches serve critical but distinct roles in a comprehensive monitoring strategy. Understanding their differences helps you decide where to invest effort and budget.
| Aspect | Real User Monitoring | Synthetic Monitoring |
|---|---|---|
| Data source | Actual user sessions | Scripted bots from fixed locations |
| Coverage | All pages users visit | Only pre-defined user flows |
| Network diversity | Real conditions (3G, 4G, Wi-Fi) | Controlled, consistent connection |
| Availability detection | Requires traffic to detect outages | Detects outages even with zero traffic |
| Baseline comparison | Harder due to variable conditions | Easy, conditions are reproducible |
| Cost model | Scales with traffic volume | Fixed cost per check interval |
The best practice is to use both: synthetic monitoring for proactive alerting and SLA tracking, and RUM for understanding true user experience.
Core Web Vitals and RUM
Google's Core Web Vitals are the primary metrics that RUM tools should capture. These metrics directly influence search rankings and correlate strongly with user satisfaction.
- Largest Contentful Paint (LCP) — measures loading performance. Target: under 2.5 seconds.
- Interaction to Next Paint (INP) — measures responsiveness. Target: under 200 milliseconds.
- Cumulative Layout Shift (CLS) — measures visual stability. Target: under 0.1.
RUM captures these metrics at the 75th percentile across all user sessions, matching the methodology Google uses for ranking signals in the Chrome User Experience Report (CrUX).
Tracking Core Web Vitals in Code
The web-vitals library provides a lightweight way to capture CWV metrics:
import { onLCP, onINP, onCLS } from 'web-vitals';
function sendToAnalytics(metric) {
const body = JSON.stringify({
name: metric.name,
value: metric.value,
rating: metric.rating,
delta: metric.delta,
id: metric.id,
navigationType: metric.navigationType
});
navigator.sendBeacon('/analytics/vitals', body);
}
onLCP(sendToAnalytics);
onINP(sendToAnalytics);
onCLS(sendToAnalytics);Implementing RUM: Step by Step
- Choose a collection strategy. Decide between a third-party RUM service or a self-hosted solution. Third-party tools like Datadog RUM, New Relic Browser, or SpeedCurve offer dashboards out of the box. Self-hosted options using the Performance API документацию give full data ownership.
- Deploy the tracking snippet. Add the RUM JavaScript to your base template. Ensure it loads asynchronously to avoid blocking the page. Place it in the
<head>for maximum coverage. - Define key pages and flows. Tag sessions with page type, user segment, and geography. This enables meaningful segmentation during analysis.
- Set up alerting thresholds. Configure alerts when the 75th percentile LCP exceeds 2.5s or INP exceeds 200ms. Alert on sudden CLS regressions that indicate layout bugs.
- Build dashboards for stakeholders. Create views for engineering (detailed timing breakdowns), product (conversion correlation), and leadership (high-level health scores).
Popular RUM Tools Compared
- Google CrUX — free, 28-day rolling data from Chrome users. Limited to origin-level granularity.
- Datadog RUM — full-featured with session replay, error tracking, and resource waterfall analysis.
- New Relic Browser — strong APM integration for correlating frontend performance with backend traces.
- SpeedCurve — focused on performance budgets and competitive benchmarking with excellent visualization.
- Sentry Performance — combines error monitoring with transaction-level performance data.
Best Practices for Production RUM
- Sample traffic at high volumes to control costs — 10-25% sampling often provides statistically significant data.
- Use
navigator.sendBeacon()for data collection to avoid data loss on page unload. - Correlate RUM data with backend APM traces using a shared trace ID header.
- Segment data by device type, connection speed, and geography to identify specific user cohorts with poor experience.
- Review RUM dashboards weekly and set up automated regression alerts for deploy-correlated performance changes.
Conclusion
Real User Monitoring bridges the gap between lab-based testing and production reality. By capturing Core Web Vitals from actual user sessions, you gain the visibility needed to prioritize performance improvements that directly impact user experience and business outcomes. Combine RUM with synthetic monitoring for a complete observability strategy that catches issues before and after they reach your users.