Lighthouse Audits
Lighthouse audits run automated Google Lighthouse tests against your website, giving you scores for performance, accessibility, SEO, and best practices. Use them to track web quality over time and catch regressions before they impact users.
How It Works
Hoppla runs a headless Lighthouse audit against your monitor's URL and records scores for each of the four audit categories. Results are stored historically so you can track trends and correlate score changes with deployments.
Lighthouse is a lab (synthetic) audit — it loads your page in a controlled headless-browser environment rather than measuring what real visitors experience. For real-user (field) Core Web Vitals collected from Chrome telemetry, Hoppla also offers a separate Core Web Vitals monitor powered by the Google CrUX dataset (LCP, INP, CLS, FCP, TTFB).
Lighthouse is an advisory check -- it produces warning status but never creates incidents. It is designed for quality tracking, not outage detection.
Configuration
| Setting | Default | Description |
|---|---|---|
| Check interval | 24 hours | Lighthouse audits are resource-intensive, so they run once a day by default. You can also trigger an audit manually from the dashboard at any time. |
| Form factor | Mobile + desktop | Each audit runs against both mobile and desktop, so you get separate scores for each. |
Results
Each audit runs and stores results for both mobile and desktop form factors. Each form factor has its own scores from 0 to 100 across the four categories below, plus its own lab metrics and audit details. The dashboard shows a mobile/desktop toggle so you can compare them.
Performance
Measures how fast your page loads and becomes interactive. Key Lighthouse lab metrics include:
- Largest Contentful Paint (LCP) -- Time until the largest content element is visible
- First Contentful Paint (FCP) -- Time until the first content is painted
- Cumulative Layout Shift (CLS) -- Visual stability as the page loads
- Total Blocking Time (TBT) -- Time the main thread is blocked during page load
These are lab measurements taken in Lighthouse's controlled environment. They approximate, but are not identical to, the field metrics in the separate Core Web Vitals (CrUX) monitor.
Accessibility
Evaluates how usable your page is for people with disabilities. Checks for proper heading hierarchy, ARIA attributes, color contrast, alt text, and keyboard navigation.
SEO
Checks fundamental search engine optimization criteria: meta tags, crawlable links, structured data, mobile viewport, and canonical URLs.
Best Practices
Evaluates general web development practices: HTTPS usage, no deprecated APIs, correct image aspect ratios, no browser errors in the console, and more.
Interpreting Scores
| Score | Rating |
|---|---|
| 90-100 | Good |
| 50-89 | Needs improvement |
| 0-49 | Poor |
Scores are color-coded in the dashboard: green for good, orange for needs improvement, and red for poor.
Alerts
Warning Notification
A warning.degraded_performance notification is sent when a Lighthouse score falls into the poor range. The notification includes the specific scores and which categories regressed.
Since Lighthouse is an advisory check, it will never create an incident or appear as downtime on status pages.
Notification Events
| Event | When |
|---|---|
warning.degraded_performance | A Lighthouse score dropped into the poor range |
warning.cleared | Scores have recovered |
Tips
- Run audits after deployments to catch performance regressions early.
- Track scores over time using the historical results to spot gradual degradation.
- Focus on LCP and CLS -- these are the Core Web Vitals that most impact user experience and search rankings.
- Do not over-optimize for Lighthouse -- a perfect 100 score is not always practical. Focus on real user impact.