What this meta tag analyzer checks

It pulls every SEO-relevant meta tag from any URL in one shot: title, meta description, canonical, viewport, language, charset, robots directives, H1, Open Graph (Facebook, LinkedIn, Slack), Twitter Card, and favicon. For each tag it tells you whether it is present, whether the length is in the optimal range, and what the value is. Three categories: core SEO, technical, social.

Most sites have one or two of these wrong. The classic patterns we see: a title that exceeds the SERP pixel limit, a missing meta description, a canonical tag pointing to the wrong URL, or Open Graph tags that mismatch the rendered title and image.

Title and description: the on-page basics

The title tag is what shows in Google search results. Aim for 30 to 60 characters and approximately 580 pixels or under to avoid truncation. The meta description sits underneath in the SERP. Aim for 120 to 160 characters. Both are critical click-through-rate levers and both get rewritten by Google when poorly matched to the query.

Missing either of these is the most common SEO mistake. The analyzer flags missing or oversized title and description fields in the issue list with severity-graded recommendations.

Canonical, viewport, and indexing directives

Every indexable page should have a self-referencing canonical tag pointing to its preferred URL. This protects against duplicate content from URL parameters, tracking codes, and trailing slash variations. Pages you do not want indexed should use the noindex meta robots tag, not robots.txt.

The viewport meta tag (content="width=device-width, initial-scale=1") is required for mobile-friendly rendering. Without it, mobile devices render the page as if it were 980 pixels wide and zoom out, breaking layout. Google's mobile-friendly test flags this immediately, and Google may demote the page.

Charset and language attributes affect how characters render and which language Google associates with the page. UTF-8 charset has been the standard for over a decade. The lang attribute on the html element (not as a meta tag) drives international SEO targeting.

Open Graph and Twitter Card tags

When users share your link on Facebook, LinkedIn, Slack, Discord, WhatsApp, or any other major platform, those services pull the Open Graph (og:*) tags to render the link preview card. Without og:title, og:description, and og:image, the shared link appears as plain text or worse, falls back to whatever scraped content the platform finds.

og:image is the highest-impact field. The recommended size is 1200x630 pixels for landscape (Facebook, LinkedIn, Slack) and at least 1200px wide for safe scaling across all platforms. Use an absolute https URL.

Twitter Card (twitter:card) is a separate tag for the Twitter/X link card style: summary, summary_large_image, app, or player. Most modern sites use summary_large_image. Twitter falls back to Open Graph when Twitter Card is missing, so og:* coverage usually gets you most of the way.

The H1 tag: technically not a meta tag, but checked here

H1 is not in the head, but it is the most important on-page element after the title. Every page should have exactly one H1 clearly stating the page's primary topic. The analyzer reports the H1 alongside meta tags because diagnosing missing H1s during a meta audit is faster than running a separate heading analysis.

For a deeper heading audit including H2 to H6 hierarchy, use our Heading Structure Analyzer. The meta tag analyzer flags missing H1 only.

How to use the results

Start with the issue list at the top. Anything flagged critical (missing title, noindex set, etc.) needs immediate fixing. Warnings are next priority and usually capture meaningful CTR or rich-result eligibility issues.

Export the JSON or CSV output to share with developers or content teams. The Markdown export works well for a quick audit summary document. Re-run after deploying fixes to confirm the changes landed correctly. Common gotcha: CMS plugins overwriting your manually set tags on save. The analyzer catches that immediately.

Why one tool beats checking tags by hand

You could open the page source and read the head yourself, but the head of a modern page is noisy. Frameworks inject preload hints, analytics snippets add their own meta tags, consent managers bury directives in script blocks, and server-rendered pages often differ from what the browser finally shows after hydration. A single analyzer normalizes all of that into one readable report so you are not scrolling through a thousand lines of minified head markup hoping to spot the one canonical tag that points at the wrong path.

It also catches the tags that are present but contradictory. A page can carry both a self-referencing canonical and a noindex directive, which sends Google a confused signal. It can declare one language in the html lang attribute and a different one in an og:locale tag. It can set a viewport tag that disables zoom, which fails accessibility audits. These conflicts are invisible when you read tags one at a time and obvious when you see them side by side.

Robots directives: the tag that silently deindexes pages

The meta robots tag is the single most dangerous tag to get wrong, because a wrong value does not break anything visible. The page loads, looks fine, and quietly drops out of Google's index. The analyzer surfaces the exact robots value so you can confirm it. The common failure is a staging-environment noindex that ships to production untouched, or a CMS template that applies noindex to an entire section by default. If a page you expect to rank is invisible in search, the robots directive is the first thing to check.

Beyond index and noindex, the robots tag carries follow, nofollow, noarchive, nosnippet, and max-snippet directives. The max-snippet and max-image-preview values now matter for AI search too, because they cap how much of your page an engine may show in a generated answer. Setting max-snippet to a tiny number to protect content can also make your page useless as an AI citation source. The analyzer shows you what you are actually telling crawlers so you can make that trade-off on purpose rather than by accident.

How meta tags feed AI search in 2026

AI answer engines read the same head you optimize for Google. The title and meta description are the cleanest summary of your page that an engine can grab without parsing the whole body, so a sharp, specific pair raises your odds of being quoted in an AI Overview, a ChatGPT answer, or a Perplexity citation. Open Graph tags feed the link cards those tools sometimes render. Schema and canonical signals tell the engine which version of a page is authoritative when several near-duplicates exist.

The practical takeaway is that meta tags are no longer just a click-through-rate lever for the ten blue links. They are part of how machines decide whether to cite you at all. A page with a vague title, a missing description, and no Open Graph image gives an AI engine almost nothing structured to work with, so it tends to favor a competitor whose head is clean and complete. Running this analyzer on your top pages is a quick way to make sure you are not handing that advantage away.

Building a repeatable meta-tag audit routine

Treat the analyzer as a recurring check, not a one-time pass. Run it on every new template before launch, because a single broken template can replicate the same missing tag across thousands of generated pages. Run it again after any platform migration, theme update, or plugin change, since those are the moments tags silently revert to defaults. For large sites, spot check one representative URL per template type rather than every page; if the product template is correct, the products usually are too.

Keep a short baseline of what each template should contain: a unique title pattern, a description pattern, a self-referencing canonical, the correct robots value, and complete Open Graph and Twitter tags. Then every audit becomes a quick comparison against that baseline instead of a fresh investigation. When you find a deviation, fix it at the template level so the correction propagates everywhere at once, and re-run to confirm. That loop, repeated on a schedule, is what keeps a growing site from slowly accumulating the small meta-tag faults that erode click-through and indexation over time.

What to do after the analyzer flags a problem

A report is only useful if it changes what you ship, so work the issue list in severity order. Missing or duplicate titles and descriptions are the fastest wins, because rewriting them is pure content work that needs no engineering. A wrong canonical or an accidental noindex is more urgent even though it is rarer, since those quietly remove a page from search entirely; fix those before you touch anything cosmetic. Social tag gaps come next, because a missing og:image costs you nothing in rankings but real attention every time the page is shared.

For each fix, change it once at the source. If a tag is wrong on one page, it is usually wrong on every page built from the same template, so correcting the template is both faster and more durable than patching pages one by one. After deploying, run the analyzer again on the same URL to confirm the value actually changed in the rendered head, not just in the CMS field, since caches and plugins can keep the old value live longer than you expect. That confirm step is what turns an audit into a real improvement instead of a list of good intentions.