Lighthouse & Core Web Vitals Audit (Bulk PageSpeed)

Run Google Lighthouse on a list of pages, on mobile and desktop, and get every score, every Core Web Vital and a ranked fix list in one table. It's the same open-source engine that powers PageSpeed Insights, but it runs on Apify, so there's no Google API key, no quota and no rate limits.

What you get for each URL and device

Field	Details
Scores (0-100)	Performance, Accessibility, Best practices, SEO
Core Web Vitals (lab)	LCP, CLS and TBT (the lab stand-in for INP), each rated good / needs-improvement / poor, plus an overall pass/fail
Other metrics	First Contentful Paint, Speed Index, server response time (TTFB), page weight, request count
Speed opportunities	Lighthouse's top fixes with estimated time and KB saved (unused JavaScript, image formats, render-blocking resources...) and the worst files
Failed checks	Accessibility, best-practice and SEO audits that failed, sorted by impact
LCP element	The HTML element that is your Largest Contentful Paint
Resources by type	Requests and KB for scripts, images, fonts, CSS and third parties

You also get a Markdown report (score table + top fixes per page) that's ready to send to a client or a dev team, and optionally the full Lighthouse HTML report for each page.

How to use it

1. Paste your URLs, one per line. Domains without https:// are fine.
2. Choose mobile, desktop or both.
3. Run. Each audit takes about 15-30 seconds.

Example output

Real run on python.org (mobile, October 2026), shortened:

{
  "url": "https://www.python.org/",
  "device": "mobile",
  "performance": 53,
  "accessibility": 75,
  "best-practices": 100,
  "seo": 83,
  "lcpMs": 4409,
  "lcpRating": "poor",
  "cls": 0,
  "clsRating": "good",
  "tbtMs": 1436,
  "tbtRating": "poor",
  "coreWebVitalsLabPass": false,
  "ttfbMs": 8,
  "totalBytes": 474662,
  "requests": 23,
  "lcpElement": "<p>",
  "opportunities": [
    {
      "title": "Eliminate render-blocking resources",
      "savingsMs": 1648,
      "savingsKb": 0,
      "topItems": [
        "https://www.python.org/static/stylesheets/style.6cb9f297dd80.css",
        "https://ajax.googleapis.com/ajax/libs/jqueryui/1.12.1/themes/smoothness/jquery-ui.css",
        "https://www.python.org/static/stylesheets/mq.ab4b4ea88926.css"
      ]
    },
    {
      "title": "Reduce unused CSS",
      "savingsMs": 760,
      "savingsKb": 129,
      "topItems": [
        "https://www.python.org/static/stylesheets/style.6cb9f297dd80.css",
        "https://www.python.org/static/stylesheets/mq.ab4b4ea88926.css"
      ]
    }
  ],
  "failedAudits": [
    {
      "category": "accessibility",
      "title": "Elements with an ARIA `[role]` that require children to contain a specific `[role]` are missing some or all of those required children.",
      "items": 3
    },
    {
      "category": "accessibility",
      "title": "`[role]`s are not contained by their required parent element",
      "items": 57
    }
  ]
}

The Markdown report starts with a table like this:

Page	Device	Perf	A11y	Best pr.	SEO	LCP	CLS	TBT	CWV (lab)
https://www.python.org/	mobile	53	75	100	83	4.4s	0	1436ms	FAIL
https://www.python.org/	desktop	69	79	100	83	1.6s	0.124	429ms	FAIL

Who it's for

• SEO agencies and freelancers: audit every client landing page at once and export the fix list.
• Developers: track performance across releases, or compare staging with production.
• Site owners: find out why Google says your site is slow, and which fixes matter most.

Pricing

Pay per audit: you're charged once per URL per device that returns a result. Failed audits (unreachable page, timeout) are free.

Notes

• These are lab measurements: a simulated mid-range phone on slow 4G (mobile) or a fast desktop connection, like Lighthouse in Chrome DevTools. Real-user (field) data from the Chrome UX Report can differ.
• Scores vary by a few points between runs, just like in PageSpeed Insights. For important pages, run twice and compare.
• Only test sites you're allowed to test. Each audit loads the page once, like a normal visitor.