PageSpeed monitoring

Fast on launch day.Fast in March, too.

PageSpeed monitors run Google Lighthouse against your pages on a schedule and keep the scores. When a deploy quietly doubles your load time, the chart says so the same day, not at the next quarterly audit.

www.example.comLighthouse · desktop
Performance
Accessibility
Best practices
SEO
Audited every 30 minutes · scores 0 to 100
4 scores

The 4 Lighthouse scores, tracked instead of sampled.

Every run records performance, accessibility, best practices and SEO, each scored 0 to 100. Green from 90 up, orange from 50 to 89, red below that, the same bands Lighthouse uses everywhere.

A one-off audit describes a single afternoon. A monitor turns the same audit into a series, and the series is what tells you whether things are drifting.

  • Performance, accessibility, best practices and SEO
  • Scored 0 to 100 on every run
  • History kept per monitor
Performance over 2 weeksregression caught
2 weeks ago1 week agotoday
The dip landed with a deploy, the fix shows in the next runs
Web vitals

Core Web Vitals, from every single run.

Each audit records Largest Contentful Paint, First Contentful Paint, Cumulative Layout Shift, Total Blocking Time and Speed Index, the metrics Google uses as ranking signals.

The targets are documented and unambiguous: LCP under 2.5 seconds, FCP under 1.8, CLS under 0.1, TBT under 200 milliseconds, Speed Index under 3.4 seconds. The monitor shows where you stand against each.

  • 5 metrics recorded per audit
  • Documented targets shown against each metric
Core Web Vitalslatest run
Largest Contentful Paint2.1 starget under 2.5 s
First Contentful Paint1.3 starget under 1.8 s
Cumulative Layout Shift0.03target under 0.1
Total Blocking Time240 mstarget under 200 ms
Speed Index2.9 starget under 3.4 s
Two strategies

Desktop and mobile, measured separately.

Each monitor audits as desktop or as mobile. Mobile runs on a simulated mid-range device over a slower network, so its scores run lower, and that lower number is what most of your visitors experience.

Create one monitor per strategy on the same URL and the gap between the two becomes a number you can watch instead of a surprise in Search Console.

  • Desktop or mobile per monitor
  • Same URL, both views, side by side
Audit strategy
DesktopMobile
Desktop performance
88
this monitor
Mobile performance
71
second monitor, same URL
Mobile scores run lower, monitor both to see the gap
Who runs this

For pages where speed pays the bills.

Ecommerce teams

Slow product pages lose buyers before the add-to-cart button renders. A scheduled audit catches the regression the day the oversized hero image ships.

SEO-focused sites

Performance is a ranking signal, and rankings move slowly enough that you want the early warning. The score moves 3 weeks before the traffic graph does.

Marketing sites after launch

Launch week gets the performance pass. Then someone embeds a video, adds a chat widget and installs one more tag. The monitor is how the launch-day speed survives the year.

Scope

What it covers, and what it doesn't.

Covered

  • Scheduled Lighthouse audits through Google's PageSpeed Insights API
  • Performance, accessibility, best practices and SEO scores
  • LCP, FCP, CLS, TBT and Speed Index per run
  • Desktop or mobile audit strategy per monitor
  • Score history and trends per page
  • Alerts on 12 notification channels

Out of scope

  • Real-user monitoring: Lighthouse is lab data, it complements field data rather than replacing it
  • Private pages: Google's API audits the URL, so it must be publicly reachable
  • Request-level profiling: for waterfalls and flame charts, use your browser's devtools
Under the hood

For the technically curious.

Google's API, your schedule

Each check calls the PageSpeed Insights v5 API with all 4 Lighthouse categories. The audit itself runs on Google's infrastructure, in a consistent environment your laptop can't replicate.

The one check that leaves home

Every other Checkmate monitor runs from your instance. PageSpeed is the exception, because a Lighthouse audit needs a real browser environment, and Google already operates it.

Bring an API key

Audits work without one, but Google throttles anonymous use. Add a free API key under Settings to raise the rate limits for steady schedules.

Quota-aware intervals

Intervals range from seconds up to 30 minutes, and every run spends API quota. Pick the longest interval that still gives you the cadence you need.

FAQ

Frequently askedquestions.

PageSpeed monitoring runs Google Lighthouse against a URL on a schedule and records the results. Checkmate keeps the 4 Lighthouse scores plus Core Web Vitals per run, so performance changes show up as a trend.

The 4 Lighthouse category scores (performance, accessibility, best practices, SEO) and 5 Core Web Vitals: Largest Contentful Paint, First Contentful Paint, Cumulative Layout Shift, Total Blocking Time and Speed Index.

Checks work without one, but Google throttles anonymous requests. A free PageSpeed Insights API key, added in Settings, raises the limits enough for regular schedules.

Both, ideally as two monitors on the same URL. Mobile audits simulate a mid-range phone on a slower network, which is closer to what most visitors experience, while desktop tracks your best case.

No. The audit runs through Google's PageSpeed Insights API, which must be able to reach the URL publicly. For availability of private pages, use an HTTP monitor instead.

Checkmate is one. It is open source under AGPL-3.0 and self-hosted, so scheduled Lighthouse audits cost you nothing beyond your own server.

Get started

Every feature,no paywall.

Checkmate is open source under AGPL-3.0. Self-host it and this feature ships free, on your servers, with your data.