TRUE PAGE LOAD MONITORING
Most tools stop when the spinner stops. We don't.
Every modern web app loads in stages. The skeleton appears, the spinner disappears — but the data hasn't arrived yet. Most monitoring tools call that "loaded." We don't.
Witnes measures until the page is genuinely usable — through SPA transitions, deferred API calls, and hydration. Because that's what your users actually experience.
THE GAP OTHER TOOLS MISS
"Our monitoring says 300ms. Users say it's slow."
Your tool measured until the skeleton appeared. Your users waited another 2 seconds for the actual data to load. Both are right — but only one matches reality.
THE GAP OTHER TOOLS MISS
"The page loads fine on the first visit. SPA navigations feel broken."
Most tools only measure full page loads. Every time a user navigates within your React or Next.js app, that transition goes unmeasured and ungraded.
HOW WE'RE DIFFERENT
What others measure vs what we measure.
The difference between a 192ms load time and a 1.66s load time is what your user actually waited for.
OTHER TOOLS
Navigation started
First Contentful Paint
Stops measuring hereSkeleton visible
Data still loading — unmeasured
Long task
Unmeasured, unreported
Page actually usable
Never recordedWITNES
Navigation started
First Contentful Paint
Noted — keeps going
Skeleton visible
Still measuring
Long task detected (85ms)
Flagged and attributed
Page actually usable
Witnes measures thisNETWORK WATERFALL
Every resource. Every millisecond.
See the full network waterfall for every page load — scripts, stylesheets, API calls. Identify exactly which resource is blocking your users without opening DevTools on their machine.
Server delivery, TTFB, FCP, DOM Interactive, LCP — all in one place, for every real visit.
TTFB
29ms
FCP
192ms
LCP
192ms
NETWORK WATERFALL (32 RESOURCES)
Long task at 1268ms (duration: 85ms)
KEY CAPABILITIES
What true load time monitoring actually includes.
Full SPA support
Every navigation inside your React, Next.js, or Vue app is tracked and graded — not just the initial page load.
Measures until usable
We wait for deferred data, API responses, and hydration to complete before recording the load time. That's what your user waited for.
Long task detection
Automatically flags tasks that blocked the main thread for over 50ms — with timestamps so engineers know exactly where to look.
Network waterfall
Every resource for every real visit — scripts, stylesheets, XHR calls — with timing breakdowns. No synthetic tests, no estimates.
Server & delivery metrics
TTFB, FCP, DOM Interactive, LCP, CLS — all captured automatically for every page load across every visitor.
Real visits only
No synthetic monitoring, no bot traffic, no lab data. Every data point comes from a real user on a real device in a real network condition.
Part of One Tool
True load time is just one layer.
01
Traffic Analytics
Where did they come from? Channel, country, device — without cookies.
02
Real User Performance Monitoring
Sentiment grades, percentiles, and long tasks — per route and per org.
You Are Here
True Page Load Monitoring
Measuring from first byte to genuinely usable — including SPA transitions.
04
Visitor Session Diagnostics
The full picture — org, route, and visit — for any specific user.
GET STARTED
Measure what your users actually waited for.
One script tag. Automatic SPA tracking. Full waterfall on every visit. No config needed.