Somewhere around 2009, the meta keywords tag died as a ranking factor, and a lot of SEO advice never caught up. Meta tags for SEO still get treated like a checklist item — stuff a keyword in, move on — when the reality is that maybe three of them actually influence anything Google does with your page. The rest either help with click-through rate, control how you look on social media, or do nothing at all. If you’re auditing a site today, you need to know which is which, because time spent perfecting a tag that has zero effect on rankings is time not spent on the tags that do.
The Meta Tags That Actually Affect Rankings
Only two meta elements have a documented, direct effect on how Google ranks or displays your page: the title tag and the meta description. Everyone calls the title tag a “meta tag” loosely, though technically it lives outside the <meta> element in the <title> tag. It’s still the single most important on-page signal you control. Google has said repeatedly that the title tag is a ranking factor, and in our own testing across client sites, rewriting a title to lead with the target keyword and a concrete benefit — a number, a year, a price — moved rankings within two to three weeks on pages that already had decent backlinks.
The meta description is different. Google confirmed years ago it’s not a ranking factor, and it’s true — you can leave it blank and rank fine. But it’s the single biggest lever you have over click-through rate, and CTR feeds into rankings indirectly through user behavior signals and, more practically, through the sheer volume of qualified traffic you pull from a fixed ranking position. A page ranking #4 with a description that states a real benefit and a call to action can out-click a #2 result with a generic auto-pulled snippet. We’ve seen CTR jump from 2.1% to 3.4% on a single description rewrite, no ranking change involved.
Title Tag Rules That Hold Up
- Keep it under 60 characters, or closer to 580 pixels if you’re being precise — Google truncates past that on desktop
- Put the focus keyword in the first 5-6 words when it reads naturally
- Every title on the site must be unique — duplicate titles across category or filter pages is the most common technical SEO issue we find in audits
- Skip the keyword-stuffed formula (“Buy Cheap Widgets | Widgets Online | Best Widgets”) — Google rewrites titles it judges to be spammy or unhelpful, and it rewrites roughly 60% of titles in some studies when the original is weak
Meta Description Best Practices That Move CTR
Write descriptions like ad copy, not like a summary. State what the searcher gets, add a specific number if you have one, and give them a reason to click your result instead of the one above it. “Learn about meta tags for SEO” tells a searcher nothing they didn’t already know from the title. “See which of the 12 meta tags Google still uses for ranking, plus copy-paste templates” gives them a reason.
Keep descriptions between 140 and 155 characters for desktop, though Google now shows up to roughly 155-160 characters on some queries and truncates unpredictably on mobile. Write the important part first — the sentence that matters should sit in the first 120 characters so it survives truncation. And don’t obsess over exact match keyword placement here; Google bolds matching terms in the snippet, but relevance to intent matters more than a literal string match.
One thing worth knowing: Google rewrites meta descriptions more often than titles — some studies put it above 70% of the time, usually pulling a snippet from on-page content instead. That’s not a reason to skip writing one. It’s a reason to make sure the content near your target keyword, usually the opening paragraph, is itself quotable as a snippet. Write your intro like it might become your meta description, because half the time it will.
Meta Tags That Do Nothing for Rankings (Stop Wasting Time Here)
The meta keywords tag is the classic example — Google confirmed in 2009 it ignores this tag entirely, and Bing has said the same. If your CMS still has a field for it, fine, fill it in for internal search or leave it blank. It changes nothing in Google Search.
The meta author tag, meta copyright tag, and meta generator tag (the one that says “WordPress 6.4”) carry no ranking weight either. Neither does meta revisit-after, a leftover from an era when crawlers needed to be told when to come back. Modern crawlers figure out your update frequency from your sitemap and your actual publishing pattern. None of these tags are harmful to have, they’re just not worth your time in an SEO audit. If you’re triaging a site with 40 issues flagged, fix the missing title tags before you touch the meta generator tag.
Robots Meta Tag: Small Tag, Big Consequences
This one gets ignored until it causes a real problem. The robots meta tag tells crawlers whether to index a page and whether to follow its links, and getting it wrong can deindex pages you need or waste crawl budget on pages you don’t.
- index, follow — the default, use it on every page you want in search results
- noindex, follow — use on thin pages you want to keep off search but still want to pass link equity through, like paginated archives or internal search results pages
- noindex, nofollow — use sparingly, mainly for staging content, thank-you pages, or admin-adjacent pages that should never appear anywhere
We’ve audited sites where a developer pushed a staging build’s noindex tag to production and the site lost 90% of its organic traffic in 11 days before anyone noticed. Check this tag on every template type — homepage, category, product, blog post — after every deploy. It takes 30 seconds and it’s the highest-consequence tag on this entire list.
Canonical Tags: Not Technically “Meta” but Just as Critical
The canonical link element lives in your <head> alongside your meta tags and gets lumped into the same conversation, so it’s worth covering here. It tells Google which version of a page is the “real” one when you have duplicates — think a product page accessible via three different URL parameters, or an HTTP and HTTPS version both live.
Self-referencing canonicals on every page are the safe default. Where teams get in trouble is e-commerce: faceted navigation (color=red&size=large) can generate thousands of near-duplicate URLs, and without canonical tags pointing back to the clean product URL, you’re splitting ranking signals across variants instead of consolidating them onto one page that can actually compete on page one.
Open Graph and Twitter Card Tags: Social, Not Search
Open Graph tags (og:title, og:description, og:image) and Twitter Card tags control how your page looks when shared on Facebook, LinkedIn, Slack, or X. They have zero direct effect on Google rankings. But skip them and your link preview shows a random image the platform scraped from your page, or worse, no image at all — and that tanks click-through when your content gets shared, which is a real traffic channel even if it’s not “SEO” in the narrow sense.
Set og:image to a 1200×630 pixel image minimum, under 5MB, and specific to the page — not one default banner reused site-wide. That single change alone typically doubles social click-through on shared links, based on what we’ve seen across client accounts running before/after tests on the same content.
How to Audit Meta Tags for SEO at Scale
Manually checking meta tags works for a 20-page site. It falls apart past a few hundred URLs. Here’s the workflow that scales:
- Crawl the site with a tool like Screaming Frog or SEOrocket’s site audit to pull every title, description, and robots tag in one export
- Filter for duplicates first — duplicate titles and descriptions are almost always a templating bug, not a content decision, and fixing the template fixes hundreds of pages at once
- Filter for missing tags — pages with no title or a blank description are usually orphaned pages, new pages that shipped without SEO review, or programmatic pages that skipped the template entirely
- Filter for length violations — titles over 60 characters and descriptions over 155 characters get flagged for rewriting, prioritized by the traffic or ranking potential of the page
- Spot-check robots tags on every unique template after any deploy, not just once a quarter
Prioritize by impact, not by tag count. A site with 500 duplicate titles on filter pages that get zero organic traffic is a lower priority than five money pages with weak, unclear titles sitting at position 6-8 on page one. Fix the pages that are one good title rewrite away from cracking the top three.
The Real Trade-Off: Optimization vs. Everything Else
Meta tags are the cheapest, fastest SEO fix available — no design changes, no dev sprint, no content rewrite, just better copy in a field you already control. That’s exactly why they’re overrated by beginners and underrated by everyone who’s been burned before. A perfect title tag on a page with 300 words of thin content and zero backlinks will not outrank a weaker title on a page that’s genuinely useful and has links pointing to it. We’ve watched thin pages with flawless meta tags sit on page three indefinitely while a messier competitor with real content and eight referring domains holds position two.
Treat meta tag optimization as maintenance, not strategy. Run the audit, fix what’s broken, move on to content depth and internal linking — because those are the levers that actually decide whether you beat the weakest page-one competitor, not the tag you spent an hour perfecting.