Synthetic monitoring tests your pages under controlled, repeatable conditions — the same device, network, and location every time. This makes it ideal for tracking performance over time, catching regressions, and diagnosing issues in detail.
This guide walks you through configuring synthetic monitoring in Calibre, from test environments through to budgets, integrations, and interpreting results. If you haven't added a Site yet, start with our Getting Started guide.
Configure test environments#
Test Profiles let you define the conditions under which your pages are tested. Each profile combines a device, network speed, and any additional settings like authentication or third-party blocking, giving you a controlled view of how your site performs across different scenarios.
Calibre automatically creates two baseline Test Profiles for every Site:
- PageSpeed Desktop
- PageSpeed Mobile
These two profiles align with Google’s PageSpeed Insights methodology, making it easy to compare your Calibre results with PageSpeed scores.
You can customise these defaults and create additional profiles to match the conditions your audience actually experiences. Mix and match device emulation, network speeds, and other settings to build a comprehensive picture of performance across the spectrum. Some ideas:
- Test with a slower network (regular 3G) to understand worst-case experiences.
- Create profiles with ads or specific third-party vendors disabled to measure their impact.
- Authenticate into your application to test logged-in experiences.
We recommend having at least one fast and one constrained profile so you can see how performance holds up across the range.
Set Performance Budgets#
Budgets are how you make sure regressions don’t go unnoticed. Set budgets against any metric available in Calibre and apply them to specific pages and Test Profiles.
When a budget is met, exceeded, or at risk, you can be notified via email or a Slack alert posted to a channel of your choice.
Create a dedicated Slack channel like #speed-metrics or #projectname-alerts so your team can keep an eye on regressions and improvements in one place.
We recommend starting with budgets for metrics that describe user experience most accurately:
- Largest Contentful Paint — how quickly the main content becomes visible.
- Cumulative Layout Shift — how much the page layout moves around unexpectedly.
- Total Blocking Time — how long the main thread is blocked from responding to user input.
- First Contentful Paint — how quickly anything first appears on screen.
- Time to First Byte — how quickly the server responds.
- Total JavaScript Transferred — the overall weight of JavaScript sent to the browser.
- Total Third Party Code Transferred — how much of that weight comes from external vendors.
You don’t need to budget everything at once. Start with a few key metrics, observe your baseline over a week or two, then tighten thresholds as you improve.
Set up Pull Request Reviews#
A majority of performance regressions are preventable if you catch them before they ship. With Pull Request Reviews, you'll see the speed impact of your work before it affects your customers.
Each review compares your production (or staging) environment against the changes in the Pull Request, highlighting differences in Core Web Vitals metrics — Largest Contentful Paint, Cumulative Layout Shift, and Total Blocking Time — along with changes in assets transferred and any performance budget breaches.
A comprehensive comparison is posted as a comment on the Pull Request, accessible to anyone using GitHub. You can enable Pull Request Reviews for any Site.
To use Pull Request Reviews, you'll need GitHub and one of the supported deployment methods: Vercel, Netlify, Heroku, or GitHub deployment statuses.
Track Deployments#
Deploy tracking adds markers to your metric graphs whenever you release, making it easy to correlate changes in performance with specific deployments. If a metric suddenly worsens, you can pinpoint exactly which release caused it.
Deploy markers appear across Synthetic monitoring views and are also available in Real User Monitoring, so you can see how a release affects both your controlled tests and your real users.
View and interpret results#
Pulse#
The Pulse page shows historical metric trends across Test Profiles, including deploy markers. You can explore 6-month, 3-month, 1-month, and 7-day averages, and reorder metric graphs to focus on what matters most.
Snapshot Overview#
Snapshot Overview provides the most detailed view of a single test run:
- Core Web Vitals and other speed metrics, colour-coded as good, needs improvement, or poor.
- A render timeline showing the visual stages of your page loading.
- A JavaScript Long Task Timeline that pinpoints blocking scripts.
- An assets-by-type graph and detailed request table for finding bottlenecks.
You can view a render video of your page and step through the timeline frame by frame, which is invaluable for understanding exactly what your users see and when.
Lighthouse audits#
The Performance, SEO, Best Practices, and Accessibility tabs surface passed and failed Lighthouse audits, each with specific recommendations. These are a practical starting point for identifying what to fix next.
Third Party analysis#
The Third Party tab breaks down how external scripts and services contribute to your overall page weight and main thread execution time. You can quickly see which vendors cost the most in terms of transfer size and processing, helping you make informed decisions about what to keep, optimise, or remove.
Connect your workflow#
Beyond Slack alerts for budgets, Calibre integrates with several tools to keep performance visible across your team:
- Slack — budget alerts, snapshot summaries, and deploy notifications.
- Webhooks — trigger custom workflows whenever Calibre events occur.
- Zapier — connect Calibre to hundreds of other tools without writing code.
- GitHub Actions — run Calibre tests as part of your CI/CD pipeline.
For a broader overview of everything Calibre offers — including CrUX and Real User Monitoring — see our Getting Started guide or explore all features.