Meta tags are HTML elements in a page’s <head> that describe the page to browsers, search engines and other software rather than displaying anything to the reader. A handful of them do real work. Most of the ones people worry about do nothing at all.
The ones worth your attention are the title tag, the meta description, the robots meta tag, and the viewport tag. Everything else is either situational or obsolete.
The title tag
Technically not a meta tag but always discussed alongside them, the <title> is the most important element in the head. It is a confirmed ranking signal, it is usually what appears as the clickable headline in the SERP, and it labels the browser tab.
What works: put the distinctive, relevant words first, since truncation happens at the end and people scan the beginning. Write for a person deciding whether to click, not for a keyword counter. Make every title on the site distinct. Include the brand where it helps recognition, usually at the end, and drop it where space is tight on a page competing hard.
On length, the practical limit is pixel width rather than character count, because Google truncates on rendered width. Roughly 55 to 60 characters is a safe working guide, but a title of narrow characters survives longer than one full of capitals and wide letters. Write the important part first and truncation stops mattering.
Google rewrites titles when it judges yours unhelpful, and has done so systematically since 2021. It draws on your H1, anchor text pointing at the page, and other on-page content. A title being rewritten is useful feedback: it usually means yours was keyword-stuffed, boilerplate, duplicated across pages, or a poor match for the query. You cannot forbid the rewriting, but writing a clear, specific, honest title makes it far less likely.
The meta description
The meta description is not a ranking factor and has not been for many years. Its job is persuasion: it frequently becomes the snippet under your listing, and a good one earns clicks that a vague one loses. Treat it as ad copy that costs nothing to place.
What works: say specifically what the page delivers and why it is worth the click, reflect the language of the query where honest, and include a reason to choose you over the listings above and below. Around 150 to 160 characters is the usual visible range, with mobile often showing less.
Google replaces descriptions often, particularly when a passage from the page answers the query better than your written summary. That is not a failure. A page that gets its description replaced by a well-chosen passage is usually being served well. It is one more argument for answer-first writing, since the passages Google picks tend to be direct answers sitting near relevant headings.
Pages without a description are not penalized; Google simply generates one. Writing them is worth the effort on pages that matter commercially and rarely worth it on hundreds of archive pages.
The robots meta tag
This one has teeth. <meta name="robots" content=".."> controls whether a page may be indexed and whether its links are followed.
index / noindex: whether the page may appear in search results. This is the correct way to keep a page out of search, as covered in indexability.
follow / nofollow: whether links on the page pass signals. noindex, follow is the usual combination for pages that should not rank but should still pass link value onward.
Snippet controls: max-snippet, nosnippet, max-image-preview and max-video-preview govern how much of your content can appear in results. These have taken on new significance because they also constrain what can be shown in AI-generated summaries, which makes them the lever publishers reach for when deciding how much of their content should be usable without a click. For most businesses that want to be found, the permissive defaults are correct.
The equivalent X-Robots-Tag HTTP header does the same job for non-HTML files such as PDFs, where there is no head section to put a tag in.
One critical interaction: a page blocked in robots.txt cannot be read, so its noindex tag is never seen. To remove a page from search, allow crawling and use noindex.
The viewport tag
<meta name="viewport" content="width=device-width, initial-scale=1"> tells mobile browsers to render at device width rather than simulating a desktop screen. Without it a responsive layout does not work and the page is judged mobile-unfriendly. It is one line, it is non-negotiable, and it is worth checking is present on every template.
Open Graph and social tags
og:title, og:description, og:image and the Twitter card equivalents control how a link looks when shared on social platforms and in messaging apps. They have no effect on search ranking, and they have a large effect on whether a shared link gets clicked. The image matters most, since a link preview with no image or a badly cropped one looks broken.
Meta keywords, and other dead weight
The keywords meta tag does nothing. Google announced it ignores the tag for ranking in 2009, and no major search engine uses it. If a plugin or an agency is still filling it in, that effort is wasted, and a public list of the terms you are targeting is a small gift to competitors.
Similarly obsolete or pointless in most cases: revisit-after, author for ranking purposes, and the various “verification” meta tags once each platform is verified another way.
Common mistakes
Duplicate titles across many pages, usually from an unmodified template. This makes pages compete and gives Google a reason to rewrite.
Keyword-stuffed titles that read like a list. These get rewritten and convert badly.
Templated descriptions repeated site-wide, which persuade nobody.
A noindex left over from staging, the single most damaging meta tag error there is, and one worth checking after every launch and database copy.
Conflicting directives, such as a meta robots tag saying one thing and an X-Robots-Tag header saying another. The more restrictive usually wins, which is rarely what anyone intended.
Missing viewport tag on a custom template.
Descriptions written for pages nobody lands on, while the commercially important pages have none.
On WordPress
SEO plugins generate titles and descriptions from templates, which is sensible for archives and adequate for most posts. The pages that earn revenue deserve hand-written ones. Yoast and Rank Math both let you override per page, and both will show you which pages are running on the default template, which is a quick way to find the gaps.
Check what is actually output in the page source rather than trusting the plugin preview, since themes occasionally emit their own competing tags.
Where this fits
Auditing and rewriting titles and descriptions at scale is part of our technical SEO services, and it is among the fastest wins available on a site that already ranks: the impressions exist, and better listings convert more of them. Our guide to title optimization, meta descriptions and HTML headings covers the craft in detail, and the effect shows up directly in click-through rate. If you have pages collecting impressions without clicks, book a discovery call.
