Lighthouse Batch Auditor — Performance, SEO & Core Web Vitals

Audit 100 URLs at once for Lighthouse performance, SEO, accessibility, and best practices scores — plus Core Web Vitals (LCP, FID, CLS). Uses Google PageSpeed Insights API..

---

Why this exists

Lighthouse is the de-facto web performance audit. This actor wraps Google's PageSpeed Insights API as a batch-friendly Apify actor with structured output for every category and every Core Web Vital, so you can audit a list of URLs in one run instead of clicking through web.dev/measure one URL at a time.

This actor:

• Batch input — audit 100s of URLs in one run
• Mobile + Desktop — run both strategies per URL, or choose one
• All 4 Lighthouse categories — performance, SEO, accessibility, best practices
• Core Web Vitals breakdown — LCP, FID, CLS, FCP, TTFB with good/needs-improvement/poor ratings
• Top optimization opportunities — actionable recommendations, prioritized by time savings
• $0.003/audit today, $0.005 from 2026-05-28 — pay-per-event, only successful audits are billed

---

Quick start

{
    "urls": ["https://example.com"],
    "strategy": "both"
}

Returns mobile and desktop scores for every URL.

---

What you get per audit

Output	Detail
Performance score	0–100 (Lighthouse scoring algorithm)
SEO score	0–100
Accessibility score	0–100
Best practices score	0–100
Core Web Vitals	LCP, FID, CLS, FCP, TTFB — each with value, unit, and good/needs-improvement/poor rating
Top optimization opportunities	Up to 5 per URL, with estimated time savings
Delta tracking	Pass in a previous run's dataset ID — output adds a delta field per URL
Mobile + Desktop	Run either or both per URL
Bring-your-own API key	For higher rate limits (25K/day)
Configurable delay between API calls	For sane rate-limit handling
Dry-run mode	Run without charges

---

How it works

This actor uses the Google PageSpeed Insights API, which runs Lighthouse in Google's infrastructure and returns the same scores you'd get from Chrome DevTools or web.dev/measure.

• No headless browser needed — pure API calls, 256MB RAM
• Free API: 25 requests per 100 seconds (without API key), 25,000/day (with API key)
• Same scores as running Lighthouse locally — Google runs the same engine

---

Input

Field	Type	Default	Description
urls	array	(required)	URLs to audit
strategy	enum	both	mobile, desktop, or both
categories	array	all 4	Which categories: performance, seo, accessibility, best-practices
googleApiKey	string	(optional)	Your Google API key for higher rate limits. Free at console.cloud.google.com
delayMs	integer	1500	Delay between API calls in ms. Lower = faster but may hit rate limits
previousRunDatasetId	string	(optional)	Dataset ID from a previous run. Output includes delta showing score changes.
dryRun	boolean	false	Run audits without charges

---

Output

Each URL × strategy combination produces one dataset item:

{
    "url": "https://example.com",
    "strategy": "mobile",
    "scores": {
        "performance": 87,
        "seo": 95,
        "accessibility": 78,
        "bestPractices": 92
    },
    "coreWebVitals": {
        "lcp": { "value": 2.1, "unit": "s", "rating": "good" },
        "fid": { "value": 45, "unit": "ms", "rating": "good" },
        "cls": { "value": 0.08, "unit": "", "rating": "needs-improvement" },
        "fcp": { "value": 1.5, "unit": "s", "rating": "good" },
        "ttfb": { "value": 320, "unit": "ms", "rating": "good" }
    },
    "topOpportunities": [
        {
            "id": "unused-css-rules",
            "title": "Remove unused CSS",
            "savings": "0.8 s",
            "description": "Remove dead rules from stylesheets..."
        }
    ],
    "auditedAt": "2026-04-13T10:00:00Z",
    "status": "success",
    "error": null
}

Delta tracking (score changes over time)

If you provide previousRunDatasetId, each result includes a delta object:

{
    "delta": {
        "performance": +5,
        "seo": 0,
        "accessibility": -2,
        "bestPractices": +3
    },
    "previousAuditedAt": "2026-04-10T10:00:00Z"
}

Find the dataset ID in the Output tab of any previous run. This lets you track whether your optimizations are actually improving scores.

Core Web Vitals ratings

Metric	Good	Needs Improvement	Poor
LCP	≤ 2.5s	2.5–4.0s	> 4.0s
FID	≤ 100ms	100–300ms	> 300ms
CLS	≤ 0.1	0.1–0.25	> 0.25

---

Pricing

$0.003 per audit (pay-per-event pricing) — rising to $0.005 per audit effective 2026-05-28 (14-day notice per Apify policy).

• "both" strategy = 2 audits per URL (mobile + desktop).
• Failed audits and dry runs are never charged.
• 100 URLs, both strategies = 200 audits = $0.60 today / $1.00 after 2026-05-28

---

Limitations

• API-based, not local Chrome. Results come from Google's PageSpeed Insights API. Scores may differ slightly from running Lighthouse locally due to Google's infrastructure.
• Rate limits without API key. Without your own Google API key, you're limited to ~25 requests per 100 seconds. For large batches, provide your own key (free to create).
• No JavaScript execution control. Unlike running Lighthouse locally, you can't customize the network throttling or CPU slowdown.

---

Related actors in this suite

Other tools by accurate_pouch for site performance + quality:

• Tech Stack Detector — 7,517 signatures across 105 categories, implies chains. $0.02/URL.
• Security Headers — 12 OWASP headers graded A+ to F, webhook. $0.004/URL.
• Broken Link Checker — Recursive crawl, sitemap + robots.txt parsing, webhook, Sheets export. $0.005/page.
• TheCrawler — Web scraper + LLM-powered structured extraction. AGPL-3.0, also on npm (thecrawler@0.1.1). $0.005/page.
• Sitemap Analyzer — Parse, validate, status-check URLs, handle sitemap indexes. $0.004/sitemap.

---

Run on Apify

No setup needed. Click above to run in the cloud. $0.003 per audit today / $0.005 after 2026-05-28.

---

Recent changes

Build 0.1.25 (2026-05-14) — Graceful handling when no Google API key is supplied: the actor now falls back to PageSpeed Insights' unauthenticated tier (~25 req / 100 s) instead of throwing. Operators can also set GOOGLE_API_KEY as an actor environment variable to apply a default key across all runs without including it per-run.