PageSpeed Insights Checker

Returns: one row per audited page and device with performanceScore, accessibilityScore, seoScore, largestContentfulPaintMs, cumulativeLayoutShift, fieldLcpP75Ms, failingAudits and sourceUrl. Input: {"urls": ["https://www.wikipedia.org/", "example.org"], "strategies": ["mobile"], "maxItems": 50} Best for: bulk Core Web Vitals audits for SEO agencies, monitoring a site's PageSpeed scores over time, and finding the slowest pages in a sitemap.

{
    "url": "https://www.wikipedia.org/",
    "strategy": "mobile",
    "performanceScore": 100,
    "seoScore": 100,
    "largestContentfulPaintMs": 1389,
    "cumulativeLayoutShift": 0.0009547550044766787,
    "fieldLcpP75Ms": 1279,
    "fieldInpP75Ms": 99
}

What does PageSpeed Insights Checker do?

PageSpeed Insights Checker runs Google PageSpeed Insights (Lighthouse) on a list of URLs, or on every page of a sitemap, and returns one clean row per page and device. Each row has the four Lighthouse category scores (performance, accessibility, best practices, SEO), the lab Core Web Vitals (LCP, CLS, TBT, FCP, Speed Index), real-user Chrome UX Report (CrUX) p75 values when Google has them, and a ranked list of the audits that fail.

You don't need a Google account, a login or an API key. By default the Actor asks the same backend the pagespeed.web.dev website uses, so you get the same reports you would see there, without pasting URLs into the website one at a time. If you already have your own free PageSpeed Insights API key, you can add it and the Actor uses the official PageSpeed Insights API instead.

It is built for jobs where one report at a time is too slow: an agency checking fifty client home pages every Monday, a web team auditing every blog post in a sitemap after a redesign, or an AI agent that needs Core Web Vitals for a handful of URLs as structured JSON.

What data can you get from PageSpeed Insights?

Field	Description	Example
url	The page URL that was audited (Lighthouse requestedUrl)	https://www.wikipedia.org/
finalUrl	The URL Lighthouse ended on after redirects	https://www.wikipedia.org/
strategy	Device profile of the report: mobile or desktop	mobile
fetchTime	When Google ran the audit, ISO 8601 UTC	2026-10-19T00:20:48.892Z
lighthouseVersion	Lighthouse version used	13.5.0
performanceScore	Performance score, 0-100	100
accessibilityScore	Accessibility score, 0-100	94
bestPracticesScore	Best-practices score, 0-100	100
seoScore	SEO score, 0-100	100
firstContentfulPaintMs	Lab First Contentful Paint, ms	1113
largestContentfulPaintMs	Lab Largest Contentful Paint, ms	1389
totalBlockingTimeMs	Lab Total Blocking Time, ms	0
cumulativeLayoutShift	Lab Cumulative Layout Shift	0.00095
speedIndexMs	Lab Speed Index, ms	2178.06
serverResponseTimeMs	Main document server response time, ms	110
totalByteWeight	Total bytes downloaded by the page	101094
largestContentfulPaintDisplay	LCP as Lighthouse shows it	1.4 s
fieldDataScope	Source of the real-user data: url, origin or null	url
fieldLcpP75Ms	Real-user LCP, 75th percentile, ms	1279
fieldInpP75Ms	Real-user Interaction to Next Paint, p75, ms	99
fieldClsP75	Real-user CLS, p75	0
fieldFcpP75Ms	Real-user First Contentful Paint, p75, ms	1156
fieldTtfbP75Ms	Real-user Time to First Byte, p75, ms	535
failingAudits	Audits scoring below 0.9, biggest savings first	see Output
runWarnings	Warnings Lighthouse raised during the run	[]
id	Stable id: strategy plus requested URL	mobile:https://www.wikipedia.org/
sourceUrl	Link to the report on pagespeed.web.dev	https://pagespeed.web.dev/analysis/https-www-wikipedia-org/w9n8erjr94?form_factor=mobile

Each entry in failingAudits has id, title, category, score, displayValue, savingsMs and savingsBytes, so you can sort pages by what to fix first.

How to use PageSpeed Insights Checker

1. Open the Actor in the Apify Console and go to the Input tab.
2. Paste your pages into Page URLs, one per line. Full URLs and bare domains such as example.org both work. Or put a sitemap.xml URL into Sitemap URLs to audit the pages it lists.
3. Choose Strategies: mobile, desktop or both. Each page gives one result per strategy.
4. Set Maximum results to cap how many reports you get (and pay for).
5. Click Start. Each analysis takes 10 to 100 seconds on Google's side; up to 3 run at the same time by default.
6. When the run finishes, open the Output tab, or export the dataset as JSON, CSV, Excel, XML or HTML.

A few things to know. maxItems counts reports, not pages. Ten pages with both strategies make up to 20 results. A page Google can't reach (a dead domain, a page that isn't HTML) is logged and skipped. It produces no result and no charge.

Real-user (CrUX) data exists only for pages and sites with enough Chrome traffic. Small sites often get null field metrics; the lab metrics and scores are still there. Lighthouse scores move a few points between runs because of network and CPU noise. For trends, schedule the Actor and compare over several runs.

Input

Field	Type	Required	Description
urls	array of strings	Yes, unless sitemapUrls is given	Pages to audit. Full URLs or bare domains (https:// is added).
sitemapUrls	array of strings	No	sitemap.xml or sitemap-index URLs; their pages are added to the list. Indexes are followed up to 2 levels deep.
strategies	array of mobile / desktop	No	Device profiles to run. Default ["mobile"].
maxItems	integer	No	Maximum number of reports, 1-10000, default 50.
maxFailingAudits	integer	No	Failing audits kept per report, 0-50, default 10.
includeFieldData	boolean	No	Add real-user CrUX p75 metrics when Google has them. Default true.
maxConcurrency	integer	No	Analyses running at the same time, 1-5, default 3.
apiKey	string (secret)	No	Your own PageSpeed Insights API key; switches to the official API.
proxyConfiguration	object	No	Proxy settings. Default: Apify Proxy (datacenter).

{
    "urls": ["https://www.wikipedia.org/", "https://www.w3.org/", "example.org"],
    "sitemapUrls": ["https://blog.apify.com/sitemap.xml"],
    "strategies": ["mobile", "desktop"],
    "maxItems": 20,
    "maxFailingAudits": 10,
    "includeFieldData": true,
    "maxConcurrency": 3,
    "proxyConfiguration": { "useApifyProxy": true }
}

Output

Real items from a local run (some failingAudits entries trimmed for length):

[
    {
        "url": "https://example.org/",
        "finalUrl": "https://example.org/",
        "strategy": "mobile",
        "fetchTime": "2026-10-19T00:20:49.213Z",
        "lighthouseVersion": "13.5.0",
        "performanceScore": 100,
        "accessibilityScore": 96,
        "bestPracticesScore": 100,
        "seoScore": 80,
        "firstContentfulPaintMs": 756,
        "largestContentfulPaintMs": 812.5,
        "totalBlockingTimeMs": 3.5,
        "cumulativeLayoutShift": 0,
        "speedIndexMs": 756,
        "serverResponseTimeMs": 1,
        "totalByteWeight": 2383,
        "largestContentfulPaintDisplay": "0.8 s",
        "fieldDataScope": "url",
        "fieldLcpP75Ms": 854,
        "fieldInpP75Ms": 71,
        "fieldClsP75": 0,
        "fieldFcpP75Ms": 755,
        "fieldTtfbP75Ms": 523,
        "failingAudits": [
            {
                "id": "landmark-one-main",
                "title": "Document does not have a main landmark.",
                "category": "accessibility",
                "score": 0,
                "displayValue": null,
                "savingsMs": null,
                "savingsBytes": null
            },
            {
                "id": "link-text",
                "title": "Links do not have descriptive text",
                "category": "seo",
                "score": 0,
                "displayValue": "1 link found",
                "savingsMs": null,
                "savingsBytes": null
            }
        ],
        "runWarnings": [],
        "id": "mobile:https://example.org/",
        "sourceUrl": "https://pagespeed.web.dev/analysis/https-example-org/75gzwciw0m?form_factor=mobile"
    },
    {
        "url": "https://www.wikipedia.org/",
        "finalUrl": "https://www.wikipedia.org/",
        "strategy": "mobile",
        "fetchTime": "2026-10-19T00:20:48.892Z",
        "lighthouseVersion": "13.5.0",
        "performanceScore": 100,
        "accessibilityScore": 94,
        "bestPracticesScore": 100,
        "seoScore": 100,
        "firstContentfulPaintMs": 1113,
        "largestContentfulPaintMs": 1389,
        "totalBlockingTimeMs": 0,
        "cumulativeLayoutShift": 0.0009547550044766787,
        "speedIndexMs": 2178.0568614394815,
        "serverResponseTimeMs": 110,
        "totalByteWeight": 101094,
        "largestContentfulPaintDisplay": "1.4 s",
        "fieldDataScope": "url",
        "fieldLcpP75Ms": 1279,
        "fieldInpP75Ms": 99,
        "fieldClsP75": 0,
        "fieldFcpP75Ms": 1156,
        "fieldTtfbP75Ms": 535,
        "failingAudits": [
            {
                "id": "cache-insight",
                "title": "Use efficient cache lifetimes",
                "category": "performance",
                "score": 0,
                "displayValue": "Est savings of 27 KiB",
                "savingsMs": 250,
                "savingsBytes": null
            },
            {
                "id": "image-delivery-insight",
                "title": "Improve image delivery",
                "category": "performance",
                "score": 0,
                "displayValue": "Est savings of 35 KiB",
                "savingsMs": 100,
                "savingsBytes": null
            }
        ],
        "runWarnings": [],
        "id": "mobile:https://www.wikipedia.org/",
        "sourceUrl": "https://pagespeed.web.dev/analysis/https-www-wikipedia-org/w9n8erjr94?form_factor=mobile"
    }
]

The overview table in the Console shows the URL, strategy, the four scores, LCP, TBT, CLS, field LCP and INP, and the report link. Open sourceUrl to see the same report on pagespeed.web.dev with screenshots and full details.

How much does it cost to check PageSpeed Insights?

PageSpeed Insights Checker is charged per result, and one result is one report (one page on one strategy). You pay only for reports returned: pages Google can't analyse cost nothing, and maxItems caps the total. See the Pricing tab for the current price.

Apify platform usage (compute and proxy) is included. If you set a maximum cost per run, the Actor stops cleanly when it is reached and tells you so in the status message.

Use with AI agents (MCP)

Connect PageSpeed Insights Checker to Claude, Cursor or any MCP client through the Apify MCP server:

{
    "mcpServers": {
        "apify": {
            "url": "https://mcp.apify.com/?tools=axiomworks/pagespeed-insights-checker"
        }
    }
}

Example prompts:

1. "Check the mobile PageSpeed score and Core Web Vitals for https://www.python.org/ and https://www.w3.org/ and tell me which one has the worse LCP."
2. "Audit the first 20 pages of https://blog.apify.com/sitemap.xml on mobile and list the five slowest by largestContentfulPaintMs, with the top failing audit for each."
3. "Compare mobile and desktop performance scores for example.org and explain the gap using the failing audits."

Agents get structured JSON back, so they can sort, compare and summarise without reading report pages.

Integrations and API

Run PageSpeed Insights Checker from your own code with the Apify API. Set APIFY_TOKEN to your Apify API token (send it in the Authorization header, never in the URL).

Python

from apify_client import ApifyClient

client = ApifyClient("<YOUR_API_TOKEN>")
run = client.actor("axiomworks/pagespeed-insights-checker").call(run_input={
    "urls": ["https://www.wikipedia.org/", "https://www.w3.org/"],
    "strategies": ["mobile", "desktop"],
    "maxItems": 50,
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["url"], item["strategy"], item["performanceScore"], item["largestContentfulPaintMs"])

JavaScript

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: '<YOUR_API_TOKEN>' });
const run = await client.actor('axiomworks/pagespeed-insights-checker').call({
    urls: ['https://www.wikipedia.org/', 'https://www.w3.org/'],
    strategies: ['mobile'],
    maxItems: 50,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

cURL

curl -X POST "https://api.apify.com/v2/acts/axiomworks~pagespeed-insights-checker/run-sync-get-dataset-items" \
  -H "Authorization: Bearer $APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"urls": ["https://www.wikipedia.org/"], "strategies": ["mobile"], "maxItems": 10}'

On the Integrations tab you can connect runs to Make, Zapier, n8n, Google Sheets, Slack or a webhook, for example to post a weekly table of scores to a sheet. Use Schedules to run the same audit every day or week and track Core Web Vitals over time.

FAQ

Do I need a Google API key? No. By default the Actor uses the backend of the public pagespeed.web.dev website, which needs no key. Google's keyless API shares one daily quota across everyone and is often exhausted, which is why the Actor doesn't rely on it. If you have your own key, put it in apiKey and the Actor calls the official API with it; if Google rejects the key, the run falls back to the default source.

Why are the field metrics null for some pages? Field data comes from the Chrome UX Report, which only covers pages and sites with enough real Chrome visitors. When Google has data for the exact page, fieldDataScope is url; when it only has data for the whole site, it is origin; when it has none, the field metrics are null.

Why do my scores differ a little from the last run? Lighthouse runs a real browser with simulated throttling, so network and CPU noise move scores by a few points. Field (CrUX) values are a 28-day rolling window and change slowly.

How long does a run take? Each analysis takes 10 to 100 seconds on Google's side. With the default 3 parallel analyses, 50 pages usually take a few minutes. You can raise maxConcurrency to 5.

Why did a URL return nothing? Google could not load it: the domain doesn't resolve, the page isn't HTML (an image or PDF), or the site blocked Lighthouse. The log names each skipped URL and the reason. Skipped URLs are not charged.

Can I audit a whole site? Yes. Put its sitemap URL into sitemapUrls. Sitemap indexes are followed up to two levels deep, and collecting stops once there are maxItems pages. Gzipped sitemaps (.xml.gz) are not supported; use the uncompressed one.

What does savingsMs mean in failingAudits? It is Lighthouse's estimate of how many milliseconds fixing that audit would save on LCP, FCP or TBT (the largest of the three). savingsBytes is the estimated bytes saved. Audits without an estimate, such as most accessibility checks, have null savings and come after the ones with savings.

Is it legal to check PageSpeed Insights?

The Actor requests public PageSpeed Insights reports, the same ones anyone can open on the pagespeed.web.dev website without logging in. It does not visit or crawl the audited pages itself, apart from reading the sitemaps you give it. You are responsible for how you use the results and for having the right to audit the URLs you submit.

PageSpeed Insights Checker is an independent tool. It is not affiliated with, endorsed by or sponsored by Google or PageSpeed Insights.

Related Actors

Actor	What it does
Sitemap URL Extractor	Lists every page URL in a sitemap, to pick which pages to audit.
Link Status Checker	Checks which links on a site are broken.
Semrush SEO Metrics Scraper	SEO traffic and keyword metrics for a domain.
Google Search SERP Scraper	Google search results for keywords, to see who ranks.

Feedback

Found a bug or need a field that isn't here? Open an issue on the Issues tab and we'll look at it.