Free metadata and structured data checker
In short
Countnest's free metadata and structured data checker reads a public page's server-rendered title, description, canonical, robots rules, Open Graph, X card tags and JSON-LD, with a score and a fix list. It does not run JavaScript or certify rich-result eligibility.
Page metadata is information in HTML tags and HTTP headers that describes a page, its preferred URL, indexing instructions and how it should appear in search and shared-link previews.
What it checks and why it matters
| What | Why it matters |
|---|---|
| Title and description | Help people and machines understand the topic; search engines may rewrite the display text. |
| Canonical and robots rules | Catch mismatched preferred URLs and accidental noindex instructions. |
| Open Graph and X cards | Help social networks show a useful title, text and image when a link is shared. |
| JSON-LD and microdata types | Flags malformed JSON and missing common properties, not every vocabulary or search eligibility rule. |
| Headings, language and viewport | Checks basic document structure, language and mobile configuration. |
How to use it, step by step
- Paste the public page URL. Use the page's clean URL; the checker removes query strings and fragments.
- Read errors first. Fix failed responses, missing titles and invalid JSON-LD before length suggestions.
- Review the share tags. Use an absolute https og:image URL and accurate page-specific titles.
- Check schema against visible content. Only describe facts visitors can see, and choose the right schema.org type.
- Validate and re-check. Use Schema.org Validator and Google's Rich Results Test for their respective checks.
Worked examples
Illustrative: missing share image
A pricing page has a good title but no og:image. Add an absolute URL to a public 1200 x 630 image, plus X large-card metadata, then refresh the social platform's cached preview.
Illustrative: invalid JSON-LD
A trailing comma makes the Article block invalid JSON. Remove it, parse the block again and use Google's Rich Results Test for supported search features.
Illustrative: wrong canonical
A product page's canonical points to the homepage. If the product is unique, use its own final URL instead. Canonicals are signals, not instructions that guarantee indexing.
Common mistakes
- One title for every page. Write a unique, useful title that matches the content.
- Relative or inaccessible preview image. Use an absolute public https URL and test its response separately.
- Length targets treated as rules. Character counts are editorial guidance. Google truncates or rewrites snippets; there is no fixed ranking limit.
- Schema that isn't visible. FAQ answers, prices and authors in the markup must agree with the page.
- Expecting FAQ and HowTo search enhancements. Google limited FAQ rich results in 2023 and stopped showing HowTo rich results. Useful markup is not a search-feature guarantee.
Key facts, with sources
- The Open Graph protocol has four required properties: og:title, og:type, og:image and og:url.
- Google recommends JSON-LD where possible for structured data.
- In 2023 Google restricted FAQ rich results to well-known government and health sites and stopped showing HowTo rich results.
- Google must be able to fetch a page to read its noindex meta tag or X-Robots-Tag header.
What a well-tagged page has
- A unique title of about 30 to 60 characters that says what the page is.
- A meta description of one or two sentences that answers the visitor's question.
- A canonical link to the preferred address, and no accidental noindex.
- One clear H1, a lang attribute and a mobile viewport.
- Open Graph title, description and image, plus an X card, so shared links show a preview.
- JSON-LD for what the page is: Organization and WebSite site-wide, BreadcrumbList on inner pages, and Article, Product, FAQPage or Event where they fit.
Metadata, structured data and AI answers
AI assistants that search the web, like ChatGPT search, Perplexity and Copilot, pick pages they can understand quickly. Structured data doesn't guarantee a citation, but it removes guesswork about the page's type, author, date and price. To see which assistants actually send you visitors, Countnest splits AI traffic by engine.
Questions and answers
What does the checker look at?
The page's title, meta description, canonical link, robots rules (meta tag and X-Robots-Tag header), language, viewport, H1 headings, Open Graph and X (Twitter) card tags, hreflang links, JSON-LD structured data and schema.org microdata.
Does it validate structured data like Google's Rich Results Test?
It parses every JSON-LD block, flags invalid JSON, lists the schema.org types it finds and checks the properties Google commonly needs for Article, Product, Organization, FAQPage, BreadcrumbList, Event, Recipe and others. Google's Rich Results Test is still the final word on rich result eligibility.
Why does metadata matter for AI search?
AI search tools such as ChatGPT search, Perplexity and Google's AI Overviews start from crawled pages. Clear titles, honest descriptions, a canonical and structured data make it easier for them to understand what a page answers and who published it.
Does it run JavaScript?
No. It reads the HTML your server sends, like most crawlers' first pass. If your tags are added by JavaScript after load, some crawlers may not see them, which is worth knowing.
Why is the query string removed from my address?
Query strings can carry personal data such as emails or IDs, so we check the page address without them. We cache the result for that address for a few hours and don't store who ran the check or their IP address.
What size should an og:image be?
1200 x 630 pixels (about 1.91:1) works across Facebook, LinkedIn, Slack and X large cards. Use an absolute https URL and keep the file under a few hundred kilobytes.
How long should a title and meta description be?
There is no fixed limit; Google truncates by pixel width. Titles of about 30 to 60 characters and descriptions of one or two sentences usually display in full.
Is this an Open Graph checker and a schema checker in one?
Yes. It checks Open Graph, X card tags, JSON-LD and microdata in the same pass as the title, description, canonical and robots rules.
Sources
- Google Search Central: Influencing your title links
- Google Search Central: Control your snippets in search results
- Google Search Central: How to specify a canonical URL
- Google Search Central: Robots meta tag and X-Robots-Tag
- Google Search Central: Introduction to structured data markup
- Google Search Central Blog: Changes to HowTo and FAQ rich results (August 2023)
- The Open Graph protocol
- X Developer Platform: Cards markup reference
- Google: Rich Results Test
- Schema.org: Schema Markup Validator
Related reading
- How to see visitors from ChatGPT and other AI assistants
- AI crawlers explained: GPTBot, ClaudeBot, PerplexityBot and how to see them
- What is AI traffic? A plain-language explainer for site owners
- Webflow: robots.txt for AI crawlers and llms.txt
- WordPress: robots.txt for AI crawlers and llms.txt
- UTM parameters for LinkedIn ads and posts
More free tools
- AI readiness checker: checks robots.txt rules for 17 AI and search crawlers, llms.txt, sitemap and homepage indexability, with a fix list.
- llms.txt generator: builds a ready-to-edit llms.txt from your homepage and sitemap, with key pages grouped, titled and summarised.
- UTM link builder: builds campaign links with utm_source, utm_medium and utm_campaign, and flags capital letters, spaces and personal data.
- Third-party tracker checker: lists the analytics, ad pixels, session recorders, embeds and font services a page loads, and the company behind each.
The metadata and structured data checker is free and stays free. Countnest analytics goes further: visits from ChatGPT, Perplexity and other assistants, AI crawler logs, search data and campaigns, without identifying visitors. 30-day free trial, no card required, then from $14 a month.
Try Countnest free for 30 days What Countnest does