Website Performance Analyzer — Speed Test & Core Web Vitals

Test the performance of any list of websites in one run. For every URL this actor measures real server response time (TTFB), full load time, page weight, compression, HTTP version, redirects, resource counts, caching and security headers — and turns them into a transparent 0–100 performance score with a graded A–F rating and the exact reason for every lost point.

Optionally, each URL is also run through the official Google PageSpeed Insights API, adding the Lighthouse performance score, lab metrics (FCP, LCP, CLS, TBT, Speed Index) and real-user Core Web Vitals field data from the Chrome UX Report.

Perfect for agencies auditing client sites, SEO monitoring, competitor benchmarking, and bulk performance checks across hundreds of URLs — no browser, no setup, results as a clean dataset you can export to JSON, CSV, or Excel.

What data you get

Field	Description
url / finalUrl	The URL you asked for and where it ended up after redirects
performanceScore	Transparent 0–100 score from direct measurements
grade	A–F rating (A ≥ 90, B ≥ 80, C ≥ 70, D ≥ 55, F below)
findings	Human-readable list of every issue that cost points
ttfbMs	Time to first byte — real server response time in ms
totalTimeMs	Time to download the full document in ms
htmlBytes	Document size (decoded)
httpVersion	HTTP/1.1, HTTP/2 … (HTTP/1.1 is flagged)
redirects	Number of redirect hops before the final URL
contentEncoding	Compression in use (gzip / br) — missing compression is flagged
resources	Counts of scripts, external scripts, stylesheets, images, iframes, approximate DOM size, viewport meta presence
caching	Cache-Control, ETag, Last-Modified, Expires headers
securityHeaders	Presence of HSTS, CSP, X-Content-Type-Options, X-Frame-Options, Referrer-Policy, Permissions-Policy
server	Server header (e.g. cloudflare, nginx)
pagespeed	Optional Google Lighthouse score, lab metrics (FCP/LCP/CLS/TBT/Speed Index) and Core Web Vitals field data

Input

{
    "urls": ["https://example.com", "https://www.wikipedia.org"],
    "strategy": "mobile",
    "runPagespeed": true
}

• urls — one URL per line; bare domains work too (https:// is added).
• strategy — mobile or desktop, used for the Lighthouse run.
• runPagespeed — set false to skip Google and rely on direct measurements only (faster).
• googleApiKey — optional, free (get one here, 25,000 requests/day). Without a key the Lighthouse part uses Google's shared anonymous quota, which is often busy — the direct measurements always work regardless.

Output sample (real run)

{
  "url": "https://example.com",
  "finalUrl": "https://example.com",
  "success": true,
  "statusCode": 200,
  "performanceScore": 98,
  "grade": "A",
  "findings": [
    "Missing Strict-Transport-Security header"
  ],
  "ttfbMs": 181,
  "totalTimeMs": 181,
  "htmlBytes": 559,
  "httpVersion": "HTTP/2",
  "redirects": 0,
  "contentType": "text/html",
  "contentEncoding": "br",
  "resources": {
    "scripts": 0,
    "externalScripts": 0,
    "stylesheets": 0,
    "images": 0,
    "iframes": 0,
    "domNodesApprox": 12,
    "hasViewportMeta": true
  },
  "caching": {
    "cacheControl": null,
    "etag": null,
    "lastModified": "Tue, 09 Jun 2026 21:01:00 GMT",
    "expires": null
  },
  "securityHeaders": {
    "strict-transport-security": false,
    "content-security-policy": false,
    "x-content-type-options": false,
    "x-frame-options": false,
    "referrer-policy": false,
    "permissions-policy": false
  },
  "server": "cloudflare"
}

A slow page looks like this instead (nytimes.com, mobile): score 69 (D) with findings such as "33 external scripts referenced (target < 15)", "Very large DOM (~4100 nodes)", and "One redirect before the final URL".

Use cases

• Agency site audits — run a client's whole sitemap through it and hand over a prioritized findings list per page.
• Competitor benchmarking — score your site against competitors on identical, objective measurements.
• Continuous monitoring — schedule it on Apify and diff scores week over week; integrate with Make/Zapier/webhooks.
• Pre-launch checks — catch missing compression, redirect chains, HTTP/1.1, missing caching or security headers before going live.

FAQ

How is the 0–100 score computed? It starts at 100 and deducts points for measurable issues (slow TTFB, missing compression, heavy HTML, HTTP/1.1, redirect chains, excessive scripts/images/DOM size, missing viewport/caching/HSTS). Every deduction appears in findings, so the score is fully explainable — no black box.

Is this the same as Lighthouse? The direct score is its own measurement of server- and document-level performance. If you enable runPagespeed, you also get the official Google Lighthouse score and Core Web Vitals alongside it.

Do I need a Google API key? No — but Lighthouse enrichment is much more reliable with one (it's free). Without a key, the actor shares Google's anonymous quota with everyone else; when that quota is busy you still get all direct measurements.

Does it render JavaScript? No — that's why it's fast and cheap. Rendering metrics (LCP, CLS, TBT) come from the optional Lighthouse integration instead.

What happens with unreachable URLs? They produce an error row (success: false) with the reason, and you are not charged for them — you pay only for successful results.

How is it priced? Pay per event: a small fee per successfully analyzed URL. Failed URLs are free.