Lighthouse Site Monitor — Core Web Vitals, Budgets & Regressions

Run Google Lighthouse on a list of pages or a whole sitemap, on mobile and desktop, and get a clean dataset of scores and Core Web Vitals. Set performance budgets to get a pass/fail per page, and schedule the Actor to catch pages that got slower since the last run.

What it does

• Bulk audits — paste URLs or give a sitemap.xml (sitemap index files are followed). Up to 500 pages per run.
• Mobile and desktop — run either or both for every page.
• Scores and metrics — Performance, Accessibility, Best Practices and SEO scores, plus LCP, FCP, TBT, CLS, Speed Index and TTI.
• Budgets with pass/fail — e.g. performance ≥ 90, lcpMs ≤ 2500, cls ≤ 0.1. Each page gets budgetPassed and a list of what failed.
• Regression detection — each run is compared with the previous run of this Actor in your account. Pages whose scores dropped or metrics got worse are listed with before/after values. Small run-to-run noise is ignored (a metric has to get worse by both the % you set and at least 200 ms, or 0.05 for CLS).
• Top opportunities — the five biggest Lighthouse savings per page, so you know what to fix first.
• Optional HTML reports — store the full Lighthouse report per page and get a link.

Typical uses

• Agencies and freelancers: audit a client's whole site in one run and export to CSV/Excel.
• Developers: check Core Web Vitals after a deploy and spot the pages that regressed.
• SEO teams: track speed and SEO scores across key landing pages every day or week.

How to monitor your site

1. Enter your sitemap URL (or a list of key pages) and set budgets.
2. Run once to create a baseline.
3. Create a Schedule in Apify (daily or weekly). Every run now reports regressions versus the run before.
4. Optional: add an integration (email, Slack, webhook) on run finish to get notified.

Input example

{
  "sitemapUrl": "https://example.com/sitemap.xml",
  "maxPages": 50,
  "devices": ["mobile", "desktop"],
  "budgets": { "performance": 90, "lcpMs": 2500, "cls": 0.1, "tbtMs": 200 },
  "compareWithPrevious": true
}

Output example (one item per page and device)

{
  "url": "https://example.com/pricing",
  "device": "mobile",
  "scores": { "performance": 72, "accessibility": 96, "bestPractices": 100, "seo": 92 },
  "metrics": { "lcpMs": 3120, "fcpMs": 1480, "tbtMs": 250, "cls": 0.12, "speedIndexMs": 2900, "ttiMs": 5100 },
  "budgetPassed": false,
  "budgetFailures": [{ "check": "lcpMs", "value": 3120, "limit": 2500, "rule": "max" }],
  "regressions": [{ "check": "performance", "previous": 85, "current": 72, "change": -13 }],
  "previousRunAt": "2026-10-04T08:00:00.000Z",
  "opportunities": [{ "id": "render-blocking-resources", "title": "Eliminate render-blocking resources", "estimatedSavingsMs": 1200 }],
  "failedAudits": { "seo": ["Document does not have a meta description"] }
}

A summary of each run (audited, budgetFailures, regressions, errors) is saved to the OUTPUT record in the key-value store.

Notes

• Results are lab data from Lighthouse running in the cloud, so absolute numbers can differ from field data (CrUX) and from your own machine. Use the same settings run to run and compare trends.
• Lighthouse scores vary a little between runs; that's why regressions use thresholds.
• Only audit sites you own or have permission to test.

Pricing

You pay per audited page and device (one mobile + one desktop audit of the same page are two audits). Failed audits are not charged.