Meta Tag Analyzer Example: Title, Description and OG Audit

The report flags og:image as missing in one table and twitter:image as missing in another, and they look like two jobs. They are one, because of a fallback that saves you half the work every time:

Twitter/X falls back to Open Graph. With no twitter:title it uses og:title; the same for description, and the same for image. This page already has twitter:card set correctly to summary_large_image — the one tag with no Open Graph equivalent, and therefore the only X-specific tag you genuinely need.

So the card is blank on X not because twitter:image is missing, but because the thing it would have fallen back to is missing too. Add og:image and both rows resolve. Adding twitter:image as well is optional, and only worth doing if you want a different image on X from the one on Facebook and LinkedIn — which is rarely true.

This warning appears on almost every audit and it deserves scepticism. Two facts, and they point the same way:

• The meta description is not a ranking factor. Keywords in it do nothing for position. It is an advertisement, not a signal.
• Google rewrites it most of the time, preferring a passage lifted from the page that better matches the specific query someone typed.

A concise, accurate 110-character description that Google actually uses beats a padded 155-character one that it discards. Fill the space only if you have something to say. Writing filler to satisfy a character target is optimising a field that is not scored, in a slot that is frequently overridden.

What the description is worth doing: having one at all. With no description, Google scrapes whatever it can find — and what it finds is often navigation text, a cookie notice, or the first line of your footer. A written description sets the floor, and it is used most reliably on the queries where your page is an obvious match, which tend to be your brand and money terms.

"Good length" is a pass on one criterion, not a guarantee the title will be used. Google generates its own title wherever it judges yours unsuitable, and the triggers have nothing to do with length:

• Boilerplate. Half your pages titled "Blog | Company" gives Google nothing to tell them apart.
• Keyword-stuffed. "SEO Guide, SEO Tips, SEO 2026, Best SEO" reads as spam and gets discarded.
• Contradicted by the page. If the H1 and the body say something else, Google resolves the conflict in favour of the content.

The replacement usually comes from your H1. That is the most actionable fact here: if you want to control the title in the SERP, make the title tag and the H1 tell the same story. A title that fights its own page will be overruled by that page.

And on the character count itself — it is a proxy. Google truncates on rendered pixel width, so a title built from narrow letters survives longer than the same character count in capitals or em-dashes. Two rules beat any target: titles cut from the right, so put the distinguishing words on the left; and put the brand last, unless the brand is the reason someone clicks.

A rewrite is not a penalty and does not affect ranking. It is Google saying your headline was not the most useful description of the page for that query — and frequently it is right.

robots: index, follow ✅

This check passes, and the tag is doing nothing. index and follow are the defaults. A page with no robots meta tag at all behaves identically. The tag is harmless, it is reassuring to see, and it is not achieving anything — which is worth knowing before you spend an afternoon adding it site-wide.

The directives that do something are the negative ones — noindex, nofollow, noarchive, nosnippet, max-snippet — and the ones you should watch for are the accidental ones. A noindex left on after a staging deploy is one of the most destructive single lines in SEO, and it will pass every check as a valid, well-formed tag.

Canonical ✅ present

Present is not the same as respected. A canonical is a hint, not a directive. Google overrides it when the rest of your signals disagree — internal links pointing elsewhere, a sitemap listing a different URL, a redirect in the way. Check the outcome in Search Console's URL Inspection, which shows "Google-selected canonical" beside "User-declared canonical". When those two differ, Google is telling you plainly that it did not believe you.

og:type: article ✅

Correct, and it unlocks properties this page is not using: article:published_time, article:modified_time, article:author and article:section. Low effort, and they give the platforms something to work with beyond a title and a picture.

og:image:width and og:image:height solve a specific, maddening problem: the first time a URL is shared, the platform has not yet fetched the image, so it renders the card without it and fills it in afterwards. Declaring the dimensions lets it reserve the space immediately. This is why your first share of a new post so often looks broken and every subsequent one looks fine — and why testing by sharing is misleading.

What is the ideal title tag length?

Google typically displays 50–60 characters of a title tag in search results. Longer titles are truncated with an ellipsis, cutting off your key message. Shorter titles may miss keyword opportunities. The Meta Tag Analyzer flags titles outside this range and shows a live character count.

Does meta description affect rankings?

Meta descriptions are not a ranking factor. They can influence click-through rate, and Google bolds terms in the snippet that match the query — but be sceptical of the further claim that a higher CTR then improves rankings through engagement signals. Google has repeatedly denied using click-through rate as a direct ranking signal, and there is no evidence that rewriting a meta description moves a position. Treat the description as an advertisement for the result, not as a lever on it — and remember Google rewrites it most of the time anyway, preferring a passage from the page that matches the specific query.

Why does og:image:alt matter?

og:image:alt provides text alternatives for your social share image, improving accessibility for users who cannot see images and helping platforms understand the image content. Facebook and LinkedIn both support og:image:alt and use it in accessibility contexts. It is also a recommended property for comprehensive Open Graph implementation.

My og:image and twitter:image are both missing. Is that two fixes?

One. Twitter/X falls back to Open Graph — with no twitter:image it uses og:image, and the same for title and description. The only tag with no Open Graph equivalent is twitter:card itself, which selects the card type, and this page already has it. Adding og:image resolves both rows. A separate twitter:image is only worth adding if you want a different picture on X from the one used elsewhere.

Should I expand my meta description to use the unused characters?

Only if you have something to say. Unused characters are not a defect: the meta description is not a ranking factor, and Google rewrites it most of the time anyway, preferring a passage from the page that matches the query. A concise, accurate description that Google uses beats a padded one it discards. Having a description at all is what matters — with none, Google scrapes whatever it finds, which is often navigation text or a cookie notice.

My title is a good length. Will Google definitely use it?

No — length is one criterion among several. Google replaces titles it judges boilerplate, keyword-stuffed, or contradicted by the page's own content, and the replacement is usually drawn from your H1. So the lever is to make the title tag and the H1 tell the same story. Note also that Google truncates on rendered pixel width rather than character count, so put the distinguishing words on the left and the brand last.

Is the "index, follow" robots tag doing anything?

No. Index and follow are the defaults — a page with no robots meta tag behaves identically. The tag is harmless and it achieves nothing. The directives that do something are the negative ones: noindex, nofollow, noarchive, nosnippet and max-snippet. Watch for the accidental ones — a noindex left on after a staging deploy is one of the most destructive single lines in SEO, and it will pass every check as a valid, well-formed tag.