Should I Noindex Tag and Category Pages?

Noindex tag pages by default, and keep category pages indexed if they're curated and genuinely useful. Tags usually generate hundreds of thin, overlapping archives nobody searches for; categories are a real navigation layer. The test is whether a human would ever choose the archive over the posts inside it.

That's the default, not a rule. Plenty of sites have one category page pulling steady traffic for a head term, and plenty have a tag system so disciplined it functions as categories under another name. The decision is per-archive, and it takes about four minutes per page to make properly.

What these pages actually are

Every blog platform generates archive URLs automatically — listings assembled from metadata rather than written by anyone. On WordPress that's four sets by default:

Archive type	What generates it	Typical count on a 200-post blog	Default call
Category	A deliberate taxonomy you defined	5–15	Index, if curated
Tag	A label added ad hoc while writing	100–400	Noindex
Date (`/2024/03/`)	Automatic, by publish month	30–60	Noindex
Author	One per contributor	1–5	Noindex on single-author sites

The count column is the whole argument. Categories are a small, intentional set. Tags multiply — most blogs end up with more tag URLs than posts, and the majority list one or two articles each. Date archives are pure duplication of content that already appears elsewhere in a better order.

Why thin archives are worth removing from the index

Four concrete costs, in the order they usually bite:

They're near-duplicates of each other. A tag listing three posts and a category listing the same three posts plus twenty more are largely the same page to a crawler. Google keeps one and quietly declines the rest — you'll see them pile up under crawled – currently not indexed in Search Console.
They compete with the posts they list. An archive for "link building" and an article titled "what is link building" target the same query, and the archive almost never wins. That's keyword cannibalization you created by accident.
They consume crawling on a site that can't spare it. On a large or slow site, 400 tag URLs is 400 requests that could have gone to articles — see what is crawl budget for when this genuinely matters and when it doesn't.
They dilute the site-level quality picture. Google's indexing decisions aren't made purely per URL. A site where most URLs are auto-generated listings gets less benefit of the doubt on the next page it crawls.

Worth saying plainly: none of this is a penalty. Indexed tag pages don't get a site punished. They just add bulk that earns nothing, and bulk has a cost.

When a category page earns its place in the index

Index it if you can answer yes to most of these:

Someone searches the term. "SEO tools" is a query. "Misc" and "Updates" are not. If nobody types the category name, there's no ranking to win.
It has content of its own. An intro passage — ideally 100–300 words of real orientation, not a keyword sentence — turns a listing into a hub page. This single change is what separates a category page that ranks from one that doesn't.
It lists enough to be useful. Ten or more posts, so the page is a genuine entry point rather than a stub.
It doesn't duplicate an article you already have. If you've written a pillar guide on the same term, that guide should own the query and the archive should be noindexed — or the two should be merged.
It already gets impressions. Filter Search Console by the archive URL. Real impressions mean Google has found a use for it, and removing it costs you something.

A category page that passes all five is essentially a topic cluster hub with a listing attached, and those are worth keeping. One that passes none is a tag page wearing a better name.

noindex, canonical, robots.txt or delete?

These four get used interchangeably and they do completely different things.

Tool	What it does	Use it for archives when
noindex, follow	Keeps the page crawlable and its links live, removes it from results	The default choice for thin archives
rel=canonical to another URL	Asks Google to consolidate two pages that are genuinely equivalent	Rarely right here — an archive isn't a duplicate of any single post
Disallow in robots.txt	Blocks crawling entirely	Almost never — see below
Delete the taxonomy	Removes the URLs completely	Tag sets that are pure noise and have no links

Two mistakes to avoid. First, don't block archives in robots.txt if you want them deindexed — Google can't read a noindex directive on a page it isn't allowed to fetch, and blocked URLs can still appear in results from links alone. Why Google indexes pages blocked in robots.txt covers exactly how that happens. Use noindex and leave the page crawlable; Google's own block indexing documentation makes the same point.

Second, the follow part isn't permanent. Google's John Mueller has said that a page left noindexed long-term eventually gets treated roughly like a 404, and its links stop being followed. For archives that's fine — they're a redundant crawl path to posts your navigation and internal links already reach. It matters only if an archive is the only route to some of your content, which is a structural problem worth fixing anyway. What are internal links in SEO covers building real paths.

Applying it

WordPress with Yoast or Rank Math: both ship a search-appearance panel with per-taxonomy toggles. Set tags, dates and author archives to "not shown in search results", then override the two or three categories you want indexed.
Next.js or another framework: export robots: { index: false, follow: true } from the archive route's metadata, and keep the exception list in code so it stays reviewable.
Either way, check the sitemap. A URL that's noindexed and still submitted in sitemap.xml sends a contradictory signal and shows up as a warning later. Whatever you deindex, remove from the sitemap in the same pass.
Verify with a fetch, not a plugin setting. Requesting the URL and reading the x-robots-tag header plus the rendered head takes ten seconds, and it catches the cases where a caching layer is serving something other than what the plugin thinks.

Expect the archives to drop out of Search Console's index over several weeks. Your reporting will get better while nothing real changes: the pages leaving were producing close to nothing.

The links question

The one genuine risk is external links. If a blogger linked to your /tag/link-building/ archive, deindexing it doesn't destroy that link's value — the page still exists and still passes signals onward while it's crawled — but you've now got equity landing on a page that can't rank. If a tag archive has earned links, that's a strong hint it should have been a real page all along. Convert it into a written hub with its own content and index it, rather than quietly noindexing something the web found useful. How to recover lost backlinks applies the same reasoning when the page is gone entirely.

That's also the wider point about archives. They were never going to bring traffic in on their own; pages people choose to link to are what do that. If you want a predictable way to earn those links, Backlinkster is a 1-for-1 exchange where site owners in related niches trade in-content links and every placement is verified live by code — free accounts get five verified swaps a month, and the plans are here.

Frequently asked questions

Should I noindex tag pages? On most blogs, yes. Tag archives are generated ad hoc, usually outnumber the posts, and typically list one or two articles each — thin, overlapping pages nobody searches for. Use noindex, follow so they stay crawlable, and keep the handful of tags that genuinely function as topic hubs.

Should category pages be noindexed too? Usually not. Categories are a small, deliberate set and often match real queries. Index a category if people search its name, it carries its own intro content, it lists ten or more posts, and it doesn't compete with an article you've already written on the same term.

Does noindexing archive pages hurt my SEO? No. It removes pages that were almost never ranking and rarely receiving traffic. The only thing to check first is whether an archive has external links or real impressions in Search Console — if it does, improve and keep it rather than deindexing it.

Should I block tag pages in robots.txt instead? No. A blocked page can't be crawled, so Google never sees the noindex directive, and the URL can still surface in results from links alone. Allow crawling and use noindex. Robots.txt controls crawling; noindex controls indexing.

Do noindexed pages still pass link value? While they're crawled, yes — noindex, follow keeps links live. But Google has said long-term noindexed pages are eventually treated like 404s and their links stop being followed, so never rely on an archive as the sole internal path to a page.

What about date and author archives? Noindex both on most sites. Date archives duplicate content that appears in better order elsewhere, and on a single-author blog the author archive is a second copy of the homepage feed. Multi-author publications with real contributor pages are the exception.

The bottom line

Default to noindex, follow on tags, dates and author archives, then keep the two or three category pages that pass the five checks — search demand, own content, enough posts, no competing article, existing impressions. Remove whatever you deindexed from the sitemap in the same pass. The work takes an afternoon, and what it buys is a smaller, cleaner set of URLs where every page has a reason to exist.