What is real user monitoring (RUM)?

Real user monitoring (RUM) measures how your site performs for actual visitors by running a small script in their browsers. It collects page loads, Core Web Vitals (LCP, INP, CLS), timings such as TTFB, JavaScript errors and the browser, device and region, so you see the experience of real users rather than that of a test machine.

Updated · by PuzariCloud · also called RUM, end-user monitoring

Why real user monitoring (RUM) matters

Synthetic checks run from data centres on fast connections; your users run older phones on mobile networks. RUM shows the real distribution, including the slow quarter of page loads that lab tests never see.

Google assesses Core Web Vitals at the 75th percentile of page loads, with good thresholds of 2.5 s for LCP, 200 ms for INP and 0.1 for CLS, so measuring p75 yourself shows where you stand.

How to measure real user monitoring (RUM)

p75 (nearest rank) = value at rank ⌈0.75 × n⌉ of the sorted samples
good at p75: LCP ≤ 2.5 s · INP ≤ 200 ms · CLS ≤ 0.1   (LCP above 4.0 s is poor)

Worked example

Eight page loads of your pricing page had LCP values of 1.2, 1.8, 2.1, 2.4, 2.6, 3.0, 3.9 and 5.2 seconds.

  1. Rank = ⌈0.75 × 8⌉ = 6.
  2. The 6th sorted value is 3.0 s, so p75 LCP = 3.0 s.
  3. 3.0 s is above 2.5 s but not above 4.0 s, so the page needs improvement, even though half the loads took 2.4 s or less.

Result: A real assessment uses far more than eight samples, but the method is the same: look at the slow quarter, not the typical load.

Common mistakes

  • Looking at averages. Web Vitals are judged at p75 for a reason. Use percentiles.
  • Mixing pages. A light landing page and a heavy dashboard in one number tell you little. Break results down by page.
  • Forgetting privacy. RUM collects data from visitors. Avoid personal data and follow your privacy and consent rules.
  • RUM instead of synthetic checks. RUM needs traffic. At night or on a new page there may be none, so keep synthetic checks for detecting outages.

How Vigil handles real user monitoring (RUM)

  • Add Vigil's script tag to your site and Real users shows Web Vitals at p75 (LCP, CLS, INP, TTFB and FCP), page views and sessions, JavaScript errors grouped by fingerprint, and browsers, devices and regions.
  • RUM sites and monthly page views are counted per plan; synthetic uptime and browser checks cover the hours without traffic.
  • Vigil's own public pages run the same script.

Real user monitoring (RUM): frequently asked questions

What is the difference between RUM and synthetic monitoring?

RUM measures real visitors as they use the site; synthetic monitoring runs scripted checks from fixed locations on a schedule. RUM shows real experience, synthetic detects outages even with no traffic.

Does RUM slow down my site?

A well-built RUM script is small and loads asynchronously, so its effect is minor but not zero. Load it with async or defer.

Which Web Vitals matter most?

The three Core Web Vitals: Largest Contentful Paint (loading), Interaction to Next Paint (responsiveness) and Cumulative Layout Shift (visual stability). TTFB and FCP help diagnose them.

Related terms

Synthetic monitoringScripted checks that act like a user on a schedule, from outside, even with no traffic. TTFB (time to first byte)Time from starting a request until the first byte of the response arrives. p95 latencyThe response time that 95% of requests beat; the slowest 5% are slower. APM (application performance monitoring)Measuring how fast and how reliably your code serves requests, down to queries and outbound calls. ObservabilityHow well you can understand a system's internal state from the data it emits.
See also: How to monitor Next.js · Vigil vs Datadog · Vigil vs New Relic
← Postmortem · All terms A–Z · SLA (service level agreement) →
Vigil by PuzariCloud

Measure it instead of estimating it

Uptime and API checks, heartbeats, status pages, on-call, traces, errors, logs and real user monitoring in one place, with every feature on every plan, including Free.

Start monitoring free Create a monitor