Meta Tags Analyzer Free Tool
Enter a URL
About Meta Tags Analyzer
What This Tool Actually Does
Meta Tags Analyzer is a server-side page reader with a narrow job. You give it a URL, and it fetches that page from this site's server in Helsinki, Finland, using a fixed Chrome 87 desktop user agent, then prints six specific pieces of metadata in a single table. It reads the title element, the meta description, the meta keywords, the meta viewport, the meta author, and it checks whether an Open Graph title exists. A seventh lookup for a site-name tag exists in the code but almost never matches anything, so the row practically never appears. Nothing else is parsed, measured, or judged.
The fetch happens once, from the server. The page you see in your own Chrome, Firefox, or Safari session is often different from what a fresh server request receives. Your browser carries cookies, logged-in state, and personalisation headers. The analyzer sends a clean request with a Chrome 87 user agent string for Windows 8.1, a Referer of https://www.google.com/, an Accept-Language of en-US, and gzip accepted. It follows redirects up to 100 hops, ignores SSL certificate errors, waits 30 seconds maximum, and uses a throwaway cookie jar. No JavaScript is executed during this fetch.
Because no JavaScript runs, any tag that a script injects after page load is invisible to the tool. Google Tag Manager containers, single-page application frameworks like React or Vue, and SEO plugins that write meta tags dynamically all fall into this category. If a page's meta description exists only after JavaScript executes, the analyzer reports "No Description" even though a human browser would show the description in the source. This is a deliberate limitation of the server-fetch design, and you should understand it before judging a JavaScript-heavy site as broken.
The output table shows the Page URL with the scheme and www stripped and the first letter capitalised, so https://www.example.com/path appears as "Example.com/path". Each row that exists prints its content. Rows that do not exist print "No Title", "No Description", or "No Keywords". The viewport row prints a dash when absent. The author row only appears when a meta author tag actually exists. The Open Graph row reads either "Open Graph meta tags is present" or "is not present". There is no colour coding, no pass or fail marking, and no score. The tool presents raw values plus two fixed advisory sentences, and it leaves interpretation to you.
A live test run for this rewrite used three addresses. The first, example.com, returned a title of "Example Domain" at 14 characters, no description, no keywords, a viewport of "width=device-width, initial-scale=1", and no Open Graph tag. The second, https://www.wikipedia.org, returned "Wikipedia" at 9 characters, a 130-character description beginning "Wikipedia is a free online encyclopedia...", no keywords, a viewport of "initial-scale=1,user-scalable=yes", and an Open Graph title present. The third, this site's own homepage, returned a 48-character title, a 135-character description, and an Open Graph title present. These three runs show the tool working as designed across a plain domain, a large international portal, and a blog.
How to Use This Tool
- Open the tool page. Find the input box above this article, labelled "Enter a URL".
- Type or paste a full address. Include the https:// prefix if the site uses HTTPS, since the tool preserves an https address as typed and only adds http:// when no scheme is present.
- Submit the form. Click the Submit button, and the server will fetch the target page with its fixed Chrome 87 user agent.
- Read the Page URL row first. This row shows the address you submitted with the scheme and www removed and the first letter capitalised.
- Inspect each metadata row. Compare the printed character counts against the advisory lines, and note which rows say "No Title", "No Description", or "No Keywords".
- Run a second site for comparison. Use the "Try New URL" link to test a competitor or a known well-optimised page, since single-page results are hard to judge without a baseline.
The whole cycle takes a few seconds per URL. The server-side fetch means you do not need to open the target site yourself, and you do not need to view source in your browser. You can test a page that blocks your country, that requires a specific user agent, or that you simply do not want to visit directly. The tool acts as an independent reader with a neutral identity.
What Each Output Row Tells You
The Meta Title row shows the exact text of the first title element in the fetched HTML, followed by the built-in advisory line "Ideally, your title tag should contain between 10 and 70 characters (Yours N characters)". The N is the character count, computed with multibyte awareness, so an accented letter like é counts as one character, not two. If the page has no title element, the row reads "No Title" and no advisory line appears.
The Meta Description row shows the content of the meta name="description" tag, followed by "Meta descriptions contains between 160 and 320 characters (Yours N characters)". Here the N counts bytes. A description full of accented Latin letters, Cyrillic, Chinese, or emoji will show a higher number than its true character length. An English description of a certain length counts as that many bytes, but a description of the same length written in accented or non-Latin letters counts noticeably higher because the tool counts bytes. This inconsistency is baked into the tool's code, and you should read description counts as byte counts, not character counts.
The Meta Keywords row prints the raw content of the meta name="keywords" tag, or "No Keywords" when the tag is absent. Google has ignored the keywords meta tag for ranking purposes since at least 2009, and Bing has stated publicly that it does not use it either. The row exists because the original vendor script that this tool derives from included it, and removing it would break the table layout. Treat this row as informational only.
The Meta Viewport row shows the content string of the meta name="viewport" tag, or a dash when absent. A typical modern value looks like "width=device-width, initial-scale=1". The viewport tag matters for mobile rendering, and its absence means a page renders at desktop width on phones. The tool does not judge the value, it only shows what is there.
The Author row appears only when a meta name="author" tag exists. Most pages do not have one, so you will rarely see this row. When it does appear, it shows the author name or organisation as written in the tag.
The Open Graph row answers one binary question. Does any meta property="og:title" tag exist in the fetched HTML? If yes, the row reads "Open Graph meta tags is present". If no, it reads "is not present". The verdict depends entirely on that single tag. A page with og:image, og:description, and og:url but no og:title will be reported as having no Open Graph tags at all.
| Row | What it reads | Missing value display |
|---|---|---|
| Page URL | Cleaned submitted address | Never missing |
| Meta Title | First title element text | "No Title" |
| Meta Description | meta name="description" content | "No Description" |
| Meta Keywords | meta name="keywords" content | "No Keywords" |
| Meta Viewport | meta name="viewport" content | "-" |
| Author | meta name="author" content | Row hidden entirely |
| Open Graph | Existence of meta property="og:title" | "is not present" |
The table has no pass or fail column. The advisory lines next to the title and description counts are static text from the original vendor script, and the numbers themselves are not compared against any threshold by the code. A 200-character title and a 5-character title both print their advisory line with the same wording. The tool shows you the raw material for your own meta tags seo analysis, and it leaves the verdict to you.
Correcting the Old Description of This Tool
The previous version of this page made claims that the live code does not support. The old copy stated that the tool reports the canonical URL, the robots meta tag with noindex and nofollow values, the page charset, language and hreflang annotations, Twitter Card tags, and og:image, og:url, and og:type. The code reads none of these. There is no canonical row, no robots row, no charset row, no hreflang row, and no Twitter Card check anywhere in the output.
The old copy also said the tool flags values as too short, too long, or missing. It does not. The code prints fixed advisory sentences with the measured number inserted, and it applies no colour, no icon, and no pass or fail judgement. A missing title prints "No Title" without any warning styling. A 400-character description prints the same neutral advisory line as a 150-character one.
The old page claimed the length thresholds reflect current search-engine limits. They do not. The title range of 10 to 70 characters and the description range of about 160 to 320 characters come from a vendor script written during Google's late-year experiment with much longer snippets, a period that predates the current search interface. Google reverted to roughly one hundred sixty characters for snippets in May 2018, and titles have been truncated by pixel width for years. The tool's advisory text has not been updated since it was written.
The old copy described the tool as scanning every recognised meta element on the page. The code reads six things you will actually see, namely the title, the description, the keywords, the viewport, the author, and whether an og:title exists, plus a site-name lookup that almost never matches. It does not scan, enumerate, or report any other meta element. The old page also claimed the tool handles JavaScript-rendered tags. It does not, since the fetch executes no JavaScript whatsoever.
| Old claim | Actual behaviour |
|---|---|
| Reports canonical URL | No canonical row exists |
| Reports robots noindex/nofollow | No robots row exists |
| Reports charset and hreflang | Neither is read |
| Reports Twitter Card tags | No Twitter Card check |
| Reports og:image, og:url, og:type | Only og:title presence is checked |
| Flags values pass or fail | Prints fixed advice, no verdict |
| Thresholds match current engines | 2017 figures, outdated since 2018 |
| Scans every meta element | Reads six specific items |
| Handles JavaScript-rendered tags | No JavaScript execution |
The correction matters because the old description set wrong expectations. Someone who ran the tool expecting a canonical URL check would find nothing and assume the tool was broken. Someone who expected a robots noindex warning would see a clean table and assume their page was indexable. The tool does what it does, and what it does is narrower than the old page advertised.
The Character Count Quirk
The title count and the description count use different units. The title count is character-based and multibyte-aware, so each Unicode code point counts as one. The description count is byte-based, so each byte in the UTF-8 encoding counts as one. For pure ASCII text, the two units are identical. For anything else, they diverge.
A description containing the word "café" counts as four bytes for café, since the é takes two bytes in UTF-8. The same word in the title counts as three characters. A Chinese description of 50 characters counts as roughly 150 bytes, because each Han character occupies three bytes in UTF-8. An emoji counts as four bytes but one character. This means a description that looks fine at 150 characters in your editor might display as 180 or 200 bytes in the tool's output.
The practical consequence is that the description advisory line can mislead you. The line says "between one hundred sixty and three hundred twenty characters", but the N it prints is a byte count. A description with one hundred sixty characters of pure ASCII prints as one hundred sixty, which sits below the stated range. The same text with ten accented characters might print as 165, which sits inside the range, even though the visible character length is identical. You cannot compare two descriptions of different languages by their byte counts and reach a fair conclusion.
For title tags, the character count is the meaningful one, since Google truncates titles by pixel width and roughly 600 pixels corresponds to about 50 to 60 characters in most fonts. The tool's ten to seventy range predates pixel-based truncation and suggests that up to seventy characters is safe. Google's display truncation typically cuts titles around 50 to 60 characters, depending on the width of the specific letters. A title of 65 narrow characters like "I" or "l" might display fully, while a title of 55 wide characters like "W" or "M" might truncate. The tool cannot know the rendered width, because it never renders the page.
| Unit | Title count | Description count |
|---|---|---|
| Counts | Characters | Bytes |
| ASCII "SEO Tools" | 9 | 9 |
| "Café" | 4 | 5 |
| 50 Chinese characters | 50 | ~150 |
| One emoji | 1 | 4 |
The byte-versus-character split is not documented anywhere on the tool page. The advisory lines both say "characters", which makes the discrepancy invisible until you test a page with non-ASCII content. If you work mainly with English-language sites, the difference never appears. If you test German, French, Spanish, or any language with accented letters, the description count runs high. If you test CJK or emoji-heavy content, the count runs very high indeed.
The Lowercasing Quirk
Before the tool parses the fetched HTML, it lowercases six words throughout the entire document. Those words are title, description, keywords, content, meta, and name. The lowercasing exists as a vendor workaround for pages that use uppercase tag names like, which some older or sloppily generated pages do. The workaround makes the parser match those tags regardless of case.
The side effect is that those six words are also lowercased inside the displayed title and description text. A page whose title contains the word "Name" will show it as "name". A page whose description mentions "Content Marketing" will show "content Marketing". The live test confirmed this with Wikipedia's article about the word "Name", whose title displayed as "name - Wikipedia" in the tool output.
This quirk corrupts displayed text in ways that can confuse you. A brand name like "Meta" in a title becomes "meta". A product called "Titleist" becomes "titleist". The tool does not restore the original case after parsing, because the lowercasing happens on the whole HTML string before any extraction. The displayed values are genuinely altered from the source.
The quirk also affects the byte count of the description, because lowercasing can change byte length in rare cases. The Turkish dotted capital İ lowercases to i with a combining dot, which changes the byte count. In practice, this edge case is vanishingly rare. The visible corruption of brand names and proper nouns is the real issue.
You should read displayed titles and descriptions with this transformation in mind. If a title looks oddly lowercased at the start of a word, check the actual page source to see the original. The tool's display is not a faithful copy of the page's metadata text. It is a lowercased approximation, faithful in content but altered in case.
Why the 2017 Length Figures Are Outdated
The advisory lines in the tool come from an earlier era of search engine behaviour. In late that year, Google experimented with search result snippets that extended to roughly three hundred twenty characters. During that window, advice about meta description length shifted upward, and the one hundred sixty to three hundred twenty character range in this tool reflects that experiment. Google reverted the experiment in May 2018, returning snippets to roughly one hundred sixty characters.
Title truncation has followed pixel width rather than character count for several years. Google's search results measure title width in pixels, and the commonly cited limit is about 600 pixels, which corresponds to roughly 50 to sixty characters in typical fonts. The exact cutoff varies with letter width, so a title of 55 wide characters can truncate while a title of 65 narrow characters displays fully. No fixed character range can capture this behaviour accurately.
The ten to seventy character range in the tool suggests that titles up to seventy characters are safe. Google's pixel-based truncation typically cuts display around 50 to sixty characters. A seventy-character title of average letter width will almost certainly truncate in search results, showing an ellipsis and cutting off the tail end of the text. The range is actively misleading for titles in the upper part of its stated span.
The description range has a similar problem. The one hundred sixty to three hundred twenty character range suggests that descriptions between those lengths are ideal. Google's current snippet display runs to roughly one hundred sixty characters, so a description of 300 characters will be cut off in most results. The useful range for a description that displays fully is closer to 120 to one hundred sixty characters, with the understanding that Google sometimes rewrites snippets or pulls text from elsewhere on the page.
| Figure | Tool's advice | Current practice |
|---|---|---|
| Title length | 10 to 70 characters | ~600 px, about 50 to 60 characters |
| Description length | 160 to 320 characters | ~160 to 160 characters |
| Snippet era | Late 2017 experiment | Reverted May 2018 |
| Truncation basis | Character count | Pixel width |
The tool's text has not been updated since the script was written. The figures are frozen in a moment when Google's behaviour was different from today's. If you use this tool as your only source of length guidance, you will aim for titles and descriptions that are longer than what Google currently displays fully. The tool remains useful for seeing what tags exist and their raw counts, but its advisory sentences should be treated as historical artefacts.
What This Tool Does Not Read
The list of things the tool ignores is longer than the list of things it reads. There is no canonical URL check, so you cannot use this tool to verify that a page points to its preferred URL version. There is no robots meta tag reading, so noindex and nofollow directives are invisible. There is no charset detection, no language or hreflang reading, and no Twitter Card validation.
The Open Graph check covers only og:title presence. The tool does not read og:image, og:description, og:url, og:type, or any other Open Graph property. A page with a complete Open Graph set of image, description, and URL but no og:title will be reported as having no Open Graph tags. A page with only og:title and nothing else will be reported as having Open Graph tags present. The verdict is binary and narrow.
Structured data in JSON-LD format is entirely ignored. Microdata and RDFa are ignored. The H1 heading is ignored. Duplicate title tags or duplicate meta descriptions are not detected. The tool reads the first title element and the first matching meta tag, and it stops there. If a page has two title elements, the tool shows only the first, and it does not warn you about the duplicate.
Social media verification tags, such as the Facebook domain verification meta tag or the Google Site Verification meta tag, are ignored. The Pinterest and LinkedIn verification tags are ignored. The theme generator meta tag, the format-detection tag, and the various mobile web app capability tags are all ignored. The tool's view of a page's metadata is deliberately minimal, and you should pair it with a full source viewer or a browser's developer tools when you need a complete picture.
| Not read | Consequence |
|---|---|
| Canonical URL | Cannot verify preferred URL |
| Robots meta | Cannot see noindex or nofollow |
| Charset | Cannot detect encoding issues |
| Hreflang | Cannot check language alternates |
| Twitter Card | Cannot validate card markup |
| og:image | Cannot check social preview image |
| JSON-LD | Cannot inspect structured data |
| H1 heading | Cannot compare title to H1 |
The tool's narrow scope is a design choice from the original script. It answers six questions about a page, and it answers them quickly and consistently. For a meta tag checker online that covers the full range of metadata, you would need a different product. For checking the basics of a title and description, this tool is sufficient.
Rate Limits and Request Behaviour
This tool has no CAPTCHA and no per-tool request cap. You can submit as many URLs as you like, one after another, without hitting a tool-specific limit. The only constraint comes from a site-wide throttle that applies to every page on limitlessreferrals.info. That throttle allows roughly fifteen requests within a single second from one visitor address. Exceeding that rate blocks the address for the rest of the day with a 503 response.
The throttle allows a standard volume of requests. Submitting a URL, reading the result, and submitting another takes several seconds at minimum. You would need to automate submissions or click very rapidly to approach fifteen requests in one second. The block lasts until the end of the day, which is a long penalty for a burst of automated testing. If you plan to check many URLs, add a small delay between submissions.
The fetch itself makes one request from this site's server to the target URL. That request carries the user agent from a Chrome build released in late 2020 and the Google Referer, which some sites treat differently from a plain browser request. Sites that block data centre IP ranges, that require JavaScript, or that serve different content to Google's crawler may return unexpected results. The tool reports what the server received, which may differ from what you see in your own browser.
The half-minute timeout handles slow sites. A page that takes longer than half a minute to respond produces the "Either input site is not valid or offline!" message. The 100-hop redirect limit handles redirect chains, though any chain that long is pathological. SSL certificate errors are ignored, so a site with an expired certificate still fetches, which is unusual for a security-conscious tool but useful for testing legacy pages.
Related Tools
When you need to build the tags this analyzer checks, use the Meta Tag Generator to create title, description, and Open Graph markup from a form.
To see how Google's crawler views a page, including rendered content and indexed text, the Google Crawler Simulator goes beyond the six tags this tool reads.
For a complete look at a page's raw HTML, including canonical, robots, and JSON-LD that this analyzer skips, the Website Source Code tool shows the full document.
Limitations
The fixed Chrome user agent dates from 2020. Chrome is well past version one hundred thirty as of this writing, and the Windows 8.1 platform string is obsolete. Most servers do not care about the user agent version, but a small number of sites serve different markup to older agents. You may see a stripped-down or compatibility version of a page that does not match what a current Chrome user receives.
The server location in Helsinki means the fetch comes from a Finnish data centre IP address. Sites that geo-target content, that block non-residential IP ranges, or that serve different versions to different countries will return results specific to Finland. A UK-only site might redirect to a holding page. A US news site might serve a cookie-consent wall. The tool reports what the Finnish server sees, which is not necessarily what your audience sees.
The lowercasing quirk alters displayed text. Brand names, proper nouns, and any word matching the six lowercased terms appear in lowercase in the output. The byte count for descriptions compounds this by counting UTF-8 bytes rather than characters. Neither behaviour is documented on the tool page, so you could mistake a display artefact for a page defect.
The Open Graph verdict is a single-tag check. A page with og:image but no og:title is reported as having no Open Graph tags, which would mislead a social media audit. The absence of a canonical check means you cannot verify duplicate content signals. The absence of a robots check means you cannot confirm indexability. The tool answers a small set of questions, and you should not generalise its output to a full technical SEO audit.
The advisory lines are frozen in that earlier period. The title range of ten to seventy characters and the description range of one hundred sixty to three hundred twenty characters do not match Google's current pixel-based title truncation or its snippet length that was established after the year following that experiment. If you follow the tool's advice literally, you will write titles and descriptions longer than what Google typically displays. The live test of this site's homepage showed a 48-character title and a 135-character description, both within the tool's stated ranges, but the description would display fully in current Google results only because it sits near the one hundred sixty character display limit.
Practical Reading of a Result Table
When you run a URL and get the table back, start by checking the Page URL row. This shows the address you submitted, cleaned of scheme and www, so you can confirm you typed the right page. A URL that redirects from http to https, from www to non-www, or through a tracking intermediary will show its final destination here. If the cleaned address differs from what you submitted, the target site has a redirect chain worth investigating.
Then look at the Meta Title row and its character count. A title under ten characters is too short to convey meaning. A title over sixty characters risks truncation in Google's pixel-based display. The tool's advisory line says up to seventy is fine, but you should aim for the practical range of 50 to sixty characters to be safe. Check that the title reads naturally and includes the primary keyword near the front.
The Meta Description row deserves a closer look because of the byte-count quirk. If the description is pure ASCII, the byte count equals the character count, and you can compare it directly to the one hundred sixty character display limit. If the description contains accented or non-Latin characters, subtract the excess bytes to estimate the true character length. A description that displays fully in Google results typically runs 120 to one hundred sixty characters, and the tool's one hundred sixty to three hundred twenty range points too high.
The Meta Keywords row is informational. Google has ignored this tag for ranking since 2009, and its presence or absence has no measurable effect on search visibility. The Meta Viewport row matters for mobile users. A missing viewport tag means the page renders at desktop width on phones, which produces the tiny-text-and-pinch-zoom experience that Google's mobile-first indexing penalises. A present viewport tag with a reasonable value like "width=device-width, initial-scale=1" is a good sign.
The Open Graph row tells you only about og:title. For social sharing, you also want og:image, og:description, and og:url, but this tool will not show them. If the row says Open Graph is present, you know at least og:title exists. If it says not present, the page lacks og:title entirely, which means link previews on Facebook, LinkedIn, and many messaging apps will fall back to whatever those platforms can guess from the page content.
| Observation | Likely meaning |
|---|---|
| Title count under 10 | Title too short for meaning |
| Title count 50 to 60 | Likely displays fully in Google |
| Title count over 70 | Likely truncates in Google |
| Description count 120 to 160 ASCII | Likely displays fully |
| Description count over 200 bytes | Byte count inflated by non-ASCII |
| Viewport present with width=device-width | Mobile rendering configured |
| Open Graph present | og:title exists |
| Open Graph not present | No og:title at all |
The numbers in the table are raw measurements, not verdicts. The tool does not tell you whether a title is good or a description is effective. It tells you what exists, how long it is, and nothing about quality. Your own judgement, informed by current search engine behaviour, has to supply the evaluation.
When to Use This Tool and When Not To
Use this tool for a quick check of the basics. You have a page, you want to know its title and description lengths, and you do not want to open the source or use browser developer tools. The tool answers in one request. Use it for comparing two competing pages side by side, for checking whether a client's pages have titles and descriptions at all, or for verifying that a freshly published page has the expected metadata.
Do not use this tool for a complete technical audit. It cannot tell you about canonical tags, robots directives, structured data, or social sharing images. Do not use it for JavaScript-heavy pages, since the server fetch skips all script execution. Do not use it to verify Google Search Console warnings about duplicate meta descriptions, since the tool does not detect duplicates. Do not use it as your source for current length guidance, since its advisory figures date from that earlier period.
The tool pairs well with the other pages in this category. Run the Meta Tag Generator to create a title and description, then run the analyzer on the resulting page to confirm the tags are present. Run the Google Crawler Simulator to see how a search engine might interpret the full page. Run the Website Source Code tool to inspect canonical, robots, and JSON-LD that the analyzer skips. Each tool covers a slice of the technical SEO picture, and the analyzer's slice is the six core meta elements.
For a meta tags seo analysis that covers the ground most people mean by that phrase, you need more than this tool offers. You need a canonical check, a robots check, an image dimension check for og:image, and a Twitter Card validator. Those exist in other products. This tool covers the title, the description, the keywords, the viewport, the author, and the og:title flag. It usually completes a check in a few seconds.
Frequently Asked Questions
What is the difference between the title character count and the description byte count?
The title count uses multibyte-aware character counting, so each Unicode code point counts as one. The description count uses raw bytes, so accented characters, CJK text, and emoji inflate the number. For pure ASCII content, the two units are identical. For any non-ASCII content, the description count runs higher than the true character count.
Why does the tool show lowercase text that is uppercase on the original page?
Before parsing, the tool lowercases the words title, description, keywords, content, meta, and name throughout the entire fetched HTML. This workaround lets it match uppercase tag names, but it also alters the displayed text. A brand name like "Meta" in a title becomes "meta" in the output. The original page source is unaffected.
Does the tool check canonical URLs, robots meta tags, or Twitter Cards?
No. The tool reads exactly six things, namely the title, meta description, meta keywords, meta viewport, meta author, and the presence of an og:title. Canonical URLs, robots noindex directives, charset declarations, hreflang annotations, Twitter Card tags, and og:image are all ignored. The previous description of this page claimed these features, and that claim was incorrect.
Are the 10 to 70 and 160 to 320 character ranges current advice?
No. Those figures date from a vendor script written during Google's experiment with longer snippets around three hundred twenty characters, an experiment that took place in the year before the current snippet format was adopted. Google reverted to roughly one hundred sixty character snippets in the year after that experiment, and titles have been truncated by pixel width, about 600 pixels or 50 to sixty characters, for years. The tool's advisory text has not been updated.
Why does the tool report "No Description" for a page that shows a description in my browser?
The tool fetches the page without executing JavaScript. If a description is injected by Google Tag Manager, a single-page application framework, or an SEO plugin after page load, the server fetch never sees it. Your browser runs the JavaScript and displays the final result. The tool reports what the raw HTML contains before any scripting.
What does the Open Graph row actually verify?
The Open Graph row checks for the existence of a single tag, meta property="og:title". If that tag exists, the row reads "Open Graph meta tags is present". If it does not, the row reads "is not present". The tool does not check og:image, og:description, og:url, or og:type. A page with those tags but no og:title is reported as having no Open Graph tags.