The Meta Tags Every Page Needs (and the Ones You Can Skip)
Open a random page's source and you'll often find twenty or thirty meta tags in the <head>. Most of them do nothing. A modern page needs about six tags to cover search engines and social platforms, and the rest are either defaults the browser already assumes or relics from search engines that shut down a decade ago.
Here's the whole working set, then the tags you can delete without a second thought.
The tags every page actually needs
Paste this into the <head> of any page and you've covered the essentials:
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>What this specific page is about</title>
<meta name="description" content="One accurate sentence describing this page.">
<link rel="canonical" href="https://example.com/this-exact-page">
Five lines. Here's why each one is there.
charset tells the browser how to decode the bytes of your HTML into characters. Without utf-8, an em dash or a curly quote can render as â€". Put it first, before the <title>, because the browser needs to know the encoding before it parses text. It's the one tag where order genuinely matters.
viewport controls how the page scales on phones. Leave it out and mobile browsers render your page at desktop width and zoom out, which fails Google's mobile-friendly check and drags down your mobile rankings. This exact string is the one you want, and you almost never need to change it.
<title> is the single most important on-page SEO element and the clickable headline in search results. It's technically not a <meta> tag, but it lives in the same place and does the same job. I won't re-cover length here; see our title tag length guide for the pixel-versus-character math.
description is the snippet Google may show under your title. It isn't a ranking factor, but it's ad copy for your organic listing. Writing one that earns clicks is its own skill, covered in how to write SEO-friendly meta descriptions.
canonical tells Google which URL is the real one when the same content is reachable at several addresses (?utm_source= variants, http versus https, trailing slashes). Get it wrong and you split your ranking signals or hide the wrong version. The canonical tags guide covers the edge cases.
Add robots only when you mean it
The robots meta tag controls whether a page is indexed and its links followed. You do not need it on normal pages. The absence of a robots tag means "index, follow," which is what you want for public content. Adding <meta name="robots" content="index, follow"> everywhere is harmless but pointless.
Reach for it when you want the opposite:
<meta name="robots" content="noindex, follow">
That keeps a page out of search results while still letting Google crawl through its links. Typical uses: thank-you pages, internal search results, thin tag archives, staging pages. If you want to keep a page out of the index, the meta tag is one way; blocking it in robots.txt is a different mechanism with a different effect, which we untangle in robots.txt vs noindex.
The Open Graph set for shareable pages
If a page will ever be shared on Slack, LinkedIn, iMessage, or X, it needs Open Graph tags so the link renders as a card instead of a bare URL:
<meta property="og:title" content="Headline for the share card">
<meta property="og:description" content="One sentence for the card.">
<meta property="og:image" content="https://example.com/card.jpg">
<meta property="og:url" content="https://example.com/this-page">
<meta property="og:type" content="website">
<meta name="twitter:card" content="summary_large_image">
The image is the part that does the work; target 1200×630 and keep it under about 1 MB. Open Graph is not a ranking factor, but the click-through you gain from a good card is real. Full walkthrough in our guide to meta OG tags, and the OG image size guide covers the dimensions.
You can generate the entire block above, filled in with your values and a live preview, with our meta tag generator. It outputs the SEO tags plus the Open Graph and Twitter set together, correctly escaped, so you're not hand-typing property="og:..." fifteen times.
Dead tags you can delete today
These show up constantly in old templates and SEO plugins. They do nothing. Delete them.
<meta name="keywords"> Google announced in 2009 that it does not use the keywords meta tag for ranking, and nothing has changed since. If anything, Bing treats stuffing it as a spam signal. The only thing this tag reliably does is hand your keyword list to competitors who view your source. Remove it.
<meta name="revisit-after" content="7 days"> This was never a standard and no major search engine has ever honored it. Googlebot decides its own crawl rate from your site's update frequency and authority. You cannot request a recrawl schedule with a meta tag.
<meta name="rating" content="general"> Meant to flag adult content for family filters. Google uses its own classifiers and SafeSearch signals, not this tag. For actual adult content there are real mechanisms; for a normal site this tag is noise.
<meta name="distribution" content="global"> A relic that supposedly told crawlers the intended distribution ("global," "local"). No search engine reads it. The web is global by default.
A few more sit in the same bucket: <meta name="author"> (fine to keep, but it's not an SEO signal), <meta name="generator"> (identifies your CMS, harmless, no SEO value), and the copyright and expires meta variants. None of them affect how you rank. If you're cleaning up a <head>, these are safe to cut.
A quick self-check
Open any page, view source (Ctrl+U or Cmd+Option+U), and confirm you see charset, viewport, title, description, and canonical. Confirm you do not see a keywords tag. If the page gets shared anywhere, confirm the six-line Open Graph block is present. That's the whole audit. For a full-page report that flags missing tags automatically, run the URL through the free SEO analyzer on our homepage.
Frequently Asked Questions
What are the most important meta tags for SEO?
The tags that carry weight are the <title>, the meta description, the viewport tag, charset, and the canonical link. The title is the strongest on-page ranking element and the description drives click-through. Viewport and charset are technical correctness. Everything beyond these is situational.
Does the meta keywords tag still work?
No. Google confirmed in 2009 that it ignores the keywords meta tag for ranking, and that remains true. It provides no benefit and can expose your target keywords to competitors. Remove it from your templates.
Do I need a robots meta tag on every page?
No. A page with no robots tag is treated as "index, follow" by default, which is what you want for public pages. Only add the tag when you need noindex to keep a page out of search results, such as thank-you or internal search pages.
How many meta tags should a page have?
Around six for a typical page: charset, viewport, title, description, canonical, and the Open Graph set if the page gets shared. There's no penalty for having a few more, but the long lists in old templates are mostly dead tags that do nothing.
What's the difference between the title tag and og:title?
The <title> is your search-result headline and an on-page ranking element. og:title is the headline shown on social share cards. They can be identical, but the social version can be punchier since it isn't competing for keyword relevance in search. Set both if a page is shareable.