Christopher Abraham home
Christopher Abraham

SEO consultant: technical SEO, indexing, schema, and AI search

Meta keywords in 2026: ignored, mostly harmless, used honestly

The meta keywords tag is the oldest SEO tactic still floating around. In the late nineties it was how you told AltaVista what a page was about, and it was abused so thoroughly that every major engine stopped trusting it. It still appears on plenty of sites, including this one, so here's the straight story.

What it is

A line in a page's head that lists the page's topics:

<meta name="keywords" content="technical SEO audit, traffic drop diagnosis">

Visitors never see it. Crawlers can read it.

What the search engines say

So the tag won't help you rank, and stuffing it can hurt. Used sparingly and honestly, it's neutral.

Why use it at all

There are a few modest reasons:

How to write one that does no harm

  1. Two to six phrases. Enough to describe the page, not enough to look like stuffing.
  2. Phrases, not single words. "Core Web Vitals" says more than "speed."
  3. Only what the page covers. Every phrase should match something in the visible content.
  4. No repetition. Don't list the same term in five variations.
  5. No competitor or brand names you don't own. It's misleading and pointless.
  6. Different on every page. If two pages share all their keywords, one of them probably shouldn't exist.

A worked example

For a guide to reading Search Console's Page indexing report, a reasonable tag reads: "Page indexing report, Search Console indexing, crawled not indexed." Three phrases, each describing a real section of the page. An unreasonable one adds "SEO, Google, best SEO, SEO expert, rank higher, cheap SEO," none of which the page is about.

Where the tag came from

Meta tags arrived with early HTML as a way for authors to describe their own documents, and the first search engines leaned on them heavily, because analyzing full page text at scale was expensive. Within a few years, site owners were filling the tag with hundreds of terms, including popular searches that had nothing to do with their pages. The engines responded by discounting the tag, then ignoring it, and moved toward signals that are harder to fake, such as links and on-page text.

Other meta tags that do matter

Should you remove it from an existing site?

Only if it's stuffed. A keywords tag with dozens of terms, or the same list on every page, is worth cleaning up, because it's the one version that might count against you. A short, honest tag can stay; removing it gains nothing.

Enforcing it on a site

On a site built with a generator, the simplest guardrail is a build check: fail the build if a page has fewer than two or more than six keywords, or repeats one. That's how this site does it, alongside checks on title and description length.

Where to put the effort instead

Everything the keywords tag once promised now comes from elsewhere: the title tag, the H1, the opening paragraph, descriptive headings, internal anchor text, and structured data. If you have an hour for SEO, spend fifty-nine minutes on those and one on the keywords tag.

For the elements that do carry weight, see on-page SEO. Terms are defined in the glossary.

Sources