Free Meta Tag Checker

If you have ever wondered how to use meta keywords for SEO, here is the honest answer: you don't. The meta keywords tag — that comma-separated list of terms you'd stuff into the page head — hasn't done anything for Google since 2009, when Google publicly confirmed it doesn't use the tag in web ranking. That's the whole story behind meta keywords 2009: the one search engine everyone cares about quietly stopped reading them and the rest followed. So when people ask do meta keywords work, or are meta keywords important, the accurate reply is that they're inert — meta keywords google support is simply gone, and Bing treats a keyword-stuffed tag as a spam signal if it treats it as anything at all.

That's also the core difference between meta tags and keywords as a concept. To define meta keywords plainly: they were one specific meta tag (meta name="keywords") meant to declare a page's topics to search engines. But meta tags as a family — title, description, canonical, robots, Open Graph — are very much alive and doing real work. The dead one is just that single keywords tag. There's no magic number for how many keywords in a meta tag you should use, because the right number is zero; the place you actually influence what a page ranks for is the visible content, the headings, and the title.

It helps to keep meta keywords vs meta description straight, since the two get confused constantly. The meta description never fed rankings either, but it does something the keywords tag never did — it shapes the snippet a searcher sees, so it still earns its place. Fussing over metadata keywords is effort better spent on a sharper title and a description worth clicking. That's why this checker reports your title, description, canonical, and robots state but doesn't grade a keywords tag: there's nothing left there to grade.

When you paste a link into Slack, iMessage, WhatsApp, or a post and it blooms into a card with a headline and a picture, that card is built from Open Graph tags. Open Graph is a small protocol Facebook introduced so any page can describe itself to social platforms, and the Open Graph protocol SEO value is indirect but real: a rich preview gets clicked and shared far more than a bare URL.

The core set is short. og:title and the og description are the headline and blurb of the card; og:url is the canonical link it points at; and og:type — written as meta property og type — tells platforms what kind of thing the page is. Most pages are simply "website", while articles, videos, and products have their own og types. og:site_name puts your brand name on the card. Get these wrong and the preview falls back to whatever the platform can scrape, which is usually ugly.

The tag that trips people up most is the image. What is an og image? It's the picture in the card, set with og:image — and og image size matters more than any other single tag here. The safe og image dimensions are 1200 by 630 pixels, a 1.91:1 ratio; go smaller and platforms may show a cramped thumbnail instead of the big card, so og:image size is the first thing to check when a preview looks wrong. Open Graph WhatsApp previews in particular are fussy and will silently drop an image that's too heavy.

Twitter/X reads Open Graph too, but layers on its own twitter cards meta tags (twitter:card, twitter:title, twitter:image) for finer control; when both are present, X prefers its own. The fastest way to see what a platform will render is a social share preview tool — X's own x card validator and a twitter card preview checker show the exact card before you publish. This tool's job is narrower and sits upstream of those: as an open graph checker it confirms the core og tags are present and populated, so you're not debugging a blank card after the link is already out.

"noindex" and "nofollow" get lumped together but do opposite jobs. A noindex tells search engines to keep a page out of the index entirely; nofollow tells them not to pass ranking credit through the links on it. You can combine them — meta noindex nofollow — on a page you want neither indexed nor used as a link source, but most of the time you want one, not both, and reaching for meta noindex nofollow by reflex is how useful pages quietly disappear.

These crawler directives live in three different places, and knowing which is which saves hours. The page-level version is a meta robots tag in the head — meta name robots content noindex is the exact line that removes a page from Google. The same instruction can arrive as an HTTP header (X-Robots-Tag), which is how an htaccess noindex rule works for non-HTML files like PDFs. And robots.txt is the third — but robots.txt no index is a trap: robots.txt controls crawling, not indexing, so a URL blocked there can still appear in results, just without a snippet. If you actually want a page gone, use the meta robots noindex, not a robots.txt no index line.

That crawl-versus-index split is the heart of robots.txt best practices: steer crawl budget away from junk (faceted URLs, internal search results) with robots.txt, and use noindex for pages that must never appear. The dangerous case is an seo noindex left on after a launch — a staging default that shipped straight to production. That's exactly what a noindex checker is for, and why this tool flags any page it fetches that's set to noindex: it's the single fastest way to catch the mistake that erases a page from Google overnight.

A canonical tag names the one URL you want indexed when the same content is reachable at several. Knowing how to use canonical tag correctly comes down to a few rules that are easy to get subtly wrong. First, every page should usually carry a self-referencing canonical pointing at itself — not just the obvious duplicates. Second, there should be exactly one: multiple canonical tags on a page cancel each other out and Google ignores all of them, so a plugin adding a second canonical on top of a hand-coded one is a common silent bug.

Some combinations actively fight each other. A canonical noindex pairing — a page that canonicalises to itself while also telling robots noindex — sends mixed signals, because a canonical says "index this one" and noindex says "index nothing"; pick one intent. And a canonical is not a redirect: the redirect vs canonical tag distinction is that a 301 moves both users and bots to the new URL, while a canonical is only a hint to bots and leaves the old page reachable. Use a redirect when the old URL should genuinely go away.

This is also where you meet Google's own phrasing. What is alternate page with proper canonical tag, the status you see in Search Console, simply means Google found the page, saw it points elsewhere via canonical, and indexed the target instead — usually working as intended, not an error. The real canonical tags SEO impact is consolidation: pointing duplicates at one master concentrates ranking signals instead of splitting them, which is the same reason duplicate title tags SEO problems hurt — scattered, competing versions of one page dilute each other. The checker shows the canonical each page declares, so you can confirm it's the URL you meant.

Every tag this tool reports is correct for the moment you ran it, and that's the limit of any on-demand check. Meta tags are among the most fragile things on a site precisely because nothing visibly breaks when they change: a CMS update rewrites a title, a redeploy ships the staging noindex, a plugin strips your canonical, an og:image starts returning a 404. The page still loads, so no error ever surfaces — the only symptom is rankings or click-through quietly sliding weeks later.

That's the gap monitoring closes. Uptura can run this same on-page check on a schedule and tell you the moment something moves — a title rewritten, a canonical removed, a page flipped to noindex, an Open Graph image gone missing — instead of you piecing it together from a traffic graph. Free during our public beta, so the SEO regression that used to cost you a month of traffic becomes something you hear about the day it happens.