SEO On-Page Audit for Any URL List

An on-page SEO report for every web page you list, as a clean table: title, meta description, headings, canonical, social tags, schema types, alt text, links, speed and a 0 to 100 score.

What does SEO On-Page Audit for Any URL List do?

You give it a list of page addresses. For each page it downloads the HTML and reads the parts that matter for on-page SEO. You get one row per page: the title and its length, the meta description, the headings, the canonical address, the Open Graph and Twitter tags, the schema.org types, how many images have no alt text, how many links stay on the site and how many leave it, the word count and how fast the server answered. The row also lists the problems it found, such as a missing title or a noindex tag, and turns them into a score from 0 to 100. Sort a few hundred pages by score and you know where to start. It works as a bulk status code, redirect and noindex checker too.

It is made for SEO agencies, in-house marketers, web developers and site owners who want a quick technical check of their own pages or their clients' pages without setting up a crawler. It audits the pages you list and nothing else: it doesn't follow links to find more pages. It reads the HTML the server sends and doesn't run JavaScript, so a page that builds its content in the browser can show few headings and words.

What data does it return?

One row for every page that was read. The same fields are always there. A field that doesn't apply to a page is empty (null).

Field	Example	Meaning
url	https://www.python.org/	The page address you listed, tidied: lower-case host name, no #fragment, and https:// added if you left it out.
finalUrl	https://www.python.org/	The address the row describes. The same as url unless redirects were followed.
status	200	The HTTP status code the page answered with.
score	84	From 0 to 100: 100 minus the points of every issue found (see the table below). Empty when the page was not audited.
issues	["title-too-short", "canonical-missing"]	The issue codes found, the most costly first. Empty when nothing is wrong.
redirects	0	How many redirects were followed to reach finalUrl.
redirectLocation	https://www.python.org/	Where the page redirects to, on a row that describes the redirect itself. Otherwise empty.
responseTimeMs	91	How long the page took to arrive, in milliseconds, measured from the machine that runs the Actor. A rough speed check, not a Core Web Vitals score.
contentLength	52966	Size of the page in bytes, after decompression.
contentType	text/html; charset=utf-8	The Content-Type the server sent.
xRobotsTag	noindex	The X-Robots-Tag header, when the server sent one.
title	Welcome to Python.org	The title tag on one line, cut at 300 characters.
titleLength	21	Length of the whole title in characters. 0 when there is none.
metaDescription	The official home of the Python Programming Language	The meta description, cut at 500 characters.
metaDescriptionLength	52	Length of the whole meta description in characters. 0 when there is none.
canonical	https://www.nasa.gov/	The canonical address the page declares, made absolute.
robotsMeta	follow, index, max-snippet:-1	The content of the robots meta tag.
viewport	width=device-width, initial-scale=1.0	The content of the viewport meta tag.
lang	en	The lang attribute of the html tag.
hreflang	[{"lang": "de", "url": "https://example.com/de/"}]	The language versions the page declares, as lang and url pairs. At most 50.
h1Count	5	Number of h1 headings that have text.
h1	["Intuitive Interpretation", "Compound Data Types"]	The text of the first 5 h1 headings, 200 characters each at most.
h2Count	9	Number of h2 headings.
h3Count	0	Number of h3 headings.
wordCount	1158	Words of visible text in the page body.
imageCount	1	Number of images.
imagesMissingAlt	0	Number of images with no alt attribute. An empty alt, which marks a decorative image, is fine.
internalLinks	126	Links that stay on the same site.
externalLinks	78	Links that go to another site.
openGraphTags	["og:type", "og:title", "og:image"]	The Open Graph tags that have a value. At most 50.
twitterTags	["twitter:card", "twitter:site"]	The Twitter card tags that have a value. At most 50.
structuredDataTypes	["WebSite", "SearchAction"]	The schema.org types found in the page's JSON-LD blocks. Only the type names are kept, not the data. At most 50.

How is the score worked out?

Every audited page starts at 100. Each issue found takes off its points, and the score never goes below 0. The issues field lists the codes, the most costly first.

Issue code	Points	It is raised when
noindex	20	A robots meta tag or the X-Robots-Tag header says noindex or none: the page asks search engines to leave it out.
title-missing	15	The page has no title, or an empty one.
description-missing	10	The page has no meta description, or an empty one.
h1-missing	10	The page has no h1 heading with text.
canonical-missing	5	The page declares no canonical address.
viewport-missing	5	The page has no viewport meta tag.
title-too-short	5	The title is under 30 characters.
title-too-long	5	The title is over 60 characters.
images-missing-alt	5	At least one image has no alt attribute.
thin-content	5	The page has fewer than 300 words of visible text.
lang-missing	3	The html tag has no lang attribute.
description-too-short	3	The meta description is under 70 characters.
description-too-long	3	The meta description is over 160 characters.
h1-multiple	3	The page has more than one h1 heading.
open-graph-missing	3	The page has none of og:title, og:description and og:image.
slow-response	3	The page took more than 2,000 ms to arrive.

Three more codes mark a page that was not audited. These rows have no score and no page-content fields.

Issue code	It is raised when
redirect	The page answered with a redirect that was not followed: followRedirects is off, there were more than 5 redirects, or the next address is not one the Actor may read (see the limits below).
http-error	The page answered with an error such as 404, 403 or 500, or with any other status that is not a success.
not-html	The page is not HTML, for example a PDF or an image.

The score is this Actor's own checklist score, not a Google metric, and it doesn't predict rankings. The length limits are common guidelines, not rules set by a search engine.

How to use it

1. Open the Input tab and add the pages to audit under Page URLs, one address per line. The prefilled list works as it is.
2. Click Start.
3. Download the results as JSON, CSV or Excel, or read them through the Apify API.

Input	Meaning
urls	The pages to audit. Each page becomes one row, in the order you list them. An address without https:// is read as https://, so example.com works. A page listed twice is audited once.
followRedirects	On by default. When a page redirects, the Actor follows up to 5 redirects and audits the page it ends on: url is the address you listed and finalUrl is where it ended. Turn it off to get the redirect itself: a row with the status code, redirectLocation and the issue redirect.
maxItems	Stop after this many results. The default is 1000. There is one result per page, so this is also the most pages a run reports.

The Actor reads about one page a second, so a list of several thousand pages needs a longer run timeout in the run options.

How much does it cost?

You pay $5.00 per 1,000 results, with no monthly fee. A page that produces a row is a result, including one that answers with an error or a redirect. A page the Actor skips (see the limits below) produces no row and costs nothing. New Apify accounts get free monthly credit, so you can try it at no cost.

Example output

A page that was audited:

{
  "url": "https://www.python.org/",
  "finalUrl": "https://www.python.org/",
  "status": 200,
  "score": 84,
  "issues": ["title-too-short", "canonical-missing", "description-too-short", "h1-multiple"],
  "redirects": 0,
  "redirectLocation": null,
  "responseTimeMs": 91,
  "contentLength": 52966,
  "contentType": "text/html; charset=utf-8",
  "xRobotsTag": null,
  "title": "Welcome to Python.org",
  "titleLength": 21,
  "metaDescription": "The official home of the Python Programming Language",
  "metaDescriptionLength": 52,
  "canonical": null,
  "robotsMeta": null,
  "viewport": "width=device-width, initial-scale=1.0",
  "lang": "en",
  "hreflang": [],
  "h1Count": 5,
  "h1": ["Intuitive Interpretation", "Compound Data Types", "All the Flow You’d Expect", "Functions Defined", "Quick & Easy to Learn"],
  "h2Count": 9,
  "h3Count": 0,
  "wordCount": 1158,
  "imageCount": 1,
  "imagesMissingAlt": 0,
  "internalLinks": 126,
  "externalLinks": 78,
  "openGraphTags": ["og:type", "og:site_name", "og:title", "og:description", "og:image", "og:image:secure_url", "og:url"],
  "twitterTags": [],
  "structuredDataTypes": ["WebSite", "SearchAction"]
}

A page that answered with an error. The page-content fields stay empty and there is no score:

{
  "url": "https://www.python.org/no-such-page-2026/",
  "finalUrl": "https://www.python.org/no-such-page-2026/",
  "status": 404,
  "score": null,
  "issues": ["http-error"],
  "redirects": 0,
  "redirectLocation": null,
  "responseTimeMs": 437,
  "contentLength": 33122,
  "contentType": "text/html; charset=utf-8",
  "xRobotsTag": null,
  "title": null,
  "titleLength": null,
  "metaDescription": null,
  "metaDescriptionLength": null,
  "canonical": null,
  "robotsMeta": null,
  "viewport": null,
  "lang": null,
  "hreflang": null,
  "h1Count": null,
  "h1": null,
  "h2Count": null,
  "h3Count": null,
  "wordCount": null,
  "imageCount": null,
  "imagesMissingAlt": null,
  "internalLinks": null,
  "externalLinks": null,
  "openGraphTags": null,
  "twitterTags": null,
  "structuredDataTypes": null
}

Data source and limits

• Where the data comes from: the pages you list, and nothing else. The Actor sends one request for each page, one for each redirect it follows and one for the robots.txt file of each site. It never calls a third-party API and never follows links to other pages.
• Only where the site allows it: before it reads a page, the Actor reads that site's robots.txt and follows the Robots Exclusion Protocol. A page the file doesn't allow for this Actor is skipped. A site with no robots.txt (a 404 or 410 answer) allows everything. If the file can't be read for any other reason (a 401, 403 or server error, or no answer), the site is treated as closed and its pages are skipped. Every redirect is checked the same way.
• Skipped pages have no row and are not charged. The run log says why each one was skipped. The reasons are: the site's robots.txt doesn't allow the page or can't be read; it asks for more than 30 seconds between requests; the address is not a valid public web address (for example it points at a private network, has a port other than 80 or 443, or contains an email address or phone number); the site keeps answering that there are too many requests (HTTP 429); the address is a file that announces more than 50 MB, such as a video or an installer; or the request fails.
• Polite pace: about one request a second in all, and a site's own Crawl-delay is kept between requests to that site. The Actor sends a User-Agent that names it.
• How fresh it is: every page is read live when the run reaches it.
• What is read: the first 5 MB of the page's HTML, as the server sends it. Content that a script adds after the page loads isn't seen.
• What is kept: the facts in the table above, not the page text. Titles, descriptions and headings are cut at the lengths shown. Email addresses and phone numbers are removed from every text the Actor keeps.
• Speed is a rough check. responseTimeMs is one download of the HTML from a data center, which can differ from what your visitors see. It leaves out images, scripts and rendering.
• Word counts are approximate. Words are counted by splitting on spaces, so pages in languages written without spaces, such as Chinese or Japanese, count low.
• Some sites turn data-center traffic away. A site's firewall can answer the Actor with an error such as 403 although a browser gets the page. You get a row with http-error. The Actor doesn't try to get around it.
• Not covered: crawling a whole site, checking that links work, page speed in a browser, content built by JavaScript, backlinks and search rankings.
• Independent tool: it isn't affiliated with, endorsed by or sponsored by any site you audit or by any search engine.

The data is public. This Actor doesn't collect personal data, and it removes email addresses and phone numbers from any text it keeps.

FAQ

Can I run it on a schedule? Yes. Use Apify Schedules to run it daily or weekly, for example to watch your key pages for a missing title or a stray noindex, and send the results to Google Sheets, Slack, a webhook or your own API.

Can I audit a whole site? Not on its own: it audits the addresses you list and doesn't follow links. Paste in the addresses from your sitemap or from a crawl you already have.

Why is a page I listed missing from the results? It was skipped, and the run log says why. The usual reasons are that the site's robots.txt doesn't allow it or can't be read, or that the site keeps answering "too many requests". The same address listed twice is audited once, and the run stops at maxItems.

Is it legal? It reads only public pages that you choose, only where the site's robots.txt allows it, at a polite rate, and keeps facts about each page rather than its text, with no personal data. Check that your own use fits each site's terms.