A category page is either your most valuable SEO landing page or a thin duplicate that Google quietly ignores. The dividing line is simple: did anyone write real content on it, or is it just an auto-generated list of post titles? This is a decision framework for sorting your category pages into “index and invest” versus “noindex or delete”, plus the exact steps to fix the ones worth keeping.
TL;DR
- Index a category page only if it has 200-300 words of original intro copy, 5+ articles, and a focused topic. Those pages can rank for broad browsing keywords your individual articles never will.
- Noindex or merge everything else. A title-plus-auto-list page duplicates the signals from the articles below it and adds nothing. At scale, those thin pages drag down site-wide quality and burn crawl budget on URLs no one searches for.
- Permanent
noindex, followis a stopgap, not a destination. Google has long said that a page left on noindex for a long time eventually behaves as if it werenoindex, nofollowand stops passing link equity. Fix or delete; do not park pages on noindex forever.
Why thin category pages hurt you
Category pages exist to group related content. To Google they land in one of two buckets:
- A real hub page with unique intro copy, curated internal links, and a clear topic. These rank for broad, browsing-intent keywords (“best X tools”, “X tutorials”) that a single deep article rarely captures.
- A near-empty list of post titles that repeats signals already present in the articles below. This is thin content by Google’s own definition: a page with no descriptive text beyond a list or grid.
Most CMS defaults (WordPress category archives, auto-generated taxonomy pages) produce the second bucket. The cost is not just that those pages don’t rank. Sites with a large number of thin, near-duplicate URLs spend crawl budget on pages no human would search for, and that pattern signals low overall quality. Google’s faceted-navigation guidance makes the same point for filter and sort URLs: pages with no unique content should be kept out of the index so crawl capacity goes to pages that matter.
How to tell which bucket a page is in
Run this quick triage on each category URL, logged out:
- The page has only an
<h1>and a list of post titles, no intro paragraph. - The Search Console Pages report lists the category URL under Crawled — currently not indexed or Discovered — currently not indexed.
site:yourdomain.com/category/x/returns the page, but the Performance report shows it earning zero clicks and near-zero impressions over the last 3 months.- The category holds fewer than 5 articles.
Hit two or more of those and the page is a noindex-or-merge candidate, not an index-and-invest one.
Index, noindex, or delete: the decision table
| Situation | Articles in category | Has unique intro copy? | Verdict |
|---|---|---|---|
| Real topic, well-developed | 5 or more | Yes (or you’ll write 200-300 words) | Index. Self-canonical, real title, meta description. |
| Real topic, no copy yet | 5 or more | No | Noindex, follow temporarily, then write copy and remove noindex. |
| Sparse, no growth planned | Fewer than 5 | Either | Merge into a sibling category; 301 the old URL. |
| Sparse but has backlinks | Fewer than 5 | Either | 301 to the merged sibling, never 404. Check the Links report first. |
| Pure filter/sort/tag view | Any | No | Noindex (or block in robots only if you also don’t need link-following). |
Step by step
- Inventory. List every category URL. The XML sitemap your CMS generates, or a crawl with a tool like Screaming Frog, gets you the full set. Visit each one as a logged-out user.
- Triage. For each “title + auto list” page, decide: write content or kill it. Fewer than 5 articles and no plan to grow it → merge into a sibling. 5+ articles representing real topic territory → keep and write copy.
- Write the intro. For keepers, write 200-300 words of original copy: what the category covers, who it’s for, and which 3-5 articles are the standouts. Vary it per category. Boilerplate you paste across every category reads as a near-duplicate pattern and gets you nowhere.
- Link out of the intro. Add contextual internal links from that intro to your 3-5 best articles in the category, with descriptive anchor text. This is the single step that converts a “duplicate list” into a hub Google treats as valuable.
- Fix the metadata. Self-referencing canonical, a meaningful
<title>(not “Posts in Category X”), and a written meta description. Make sure the category sits in your breadcrumb trail. - Handle the rejects. For pages you’re not keeping, set
<meta name="robots" content="noindex, follow">. Thefollowkeeps Google crawling the article links during the transition. Do not also block the URL inrobots.txt— Google has to crawl the page to even see the noindex tag, so a robots.txt block prevents the directive from ever being read. - Recheck at 30 days. Reopen the Pages report. Categories you wrote real content for should move from “Crawled — currently not indexed” to Indexed. If they don’t after a couple of crawl cycles, the copy is still too thin or too close to the articles below it.
The “noindex, follow” caveat most guides skip
noindex, follow is the right setting while you’re transitioning a page out — it stops the page ranking while Google still walks the internal links on it. But it is not a permanent home. Google’s John Mueller has said for years that when a page carries noindex long term, Google reads that as “this page really shouldn’t be in Search”, recrawls it less, and effectively stops following its links — so over time noindex, follow behaves like noindex, nofollow. The practical takeaway: any category you intend to keep around as a navigation hub should earn its way back to indexed. Pages you want permanently out should be merged and 301’d, not parked on noindex indefinitely.
Common pitfalls
- Noindexing every category page as a blanket rule. A well-developed category can rank for broad terms your individual articles can’t reach. Decide per page, not per template.
- Pasting the same 100-word intro everywhere to “make them unique”. Google detects the near-duplicate pattern and you’ve gained nothing.
- Footer mega-menus that link category-to-category with identical anchor text. Internal links lose value when the anchor never varies; descriptive, varied anchors carry far more signal.
- Killing a category that has external backlinks. Check Search Console → Links first. If a category URL has inbound links, 301 it to the merged sibling. A 404 throws that link equity away.
- Confusing category pages with tag pages. A category is a hub (a place); a tag is a cross-cutting filter. They’re handled differently — see the tag-pages guide below.
Who this is for
Content-site owners with roughly 5-50 category pages auditing site-wide quality. If your Pages report is full of category URLs marked “Crawled — currently not indexed” and you can’t tell whether to fix them or kill them, this framework is the sort.
When to skip this
Sites with no category structure (a single chronological feed) have nothing to audit here. Brand-new sites with no Search Console data yet should write articles first and audit categories around month 3, once there’s crawl and impression data to act on.
FAQ
- How much intro copy is “enough” for a category page?: 200-300 words of original copy that says what the category is, who it’s for, and which articles stand out. Less reads as thin; much more and you’ve written an article instead of a hub.
- Should I link from articles back to their category page?: Yes. A breadcrumb plus a “back to category” link both pass internal-linking signal and help readers. Just don’t stuff the link five times into the body.
- Can a category page and an article in it rank for the same keyword?: They can compete (keyword cannibalization). Split by intent: broad/browsing → category page; specific question → article. Give them distinct titles and link each to the other.
- What about empty categories I plan to fill later?: Noindex, follow now; write the articles; remove the noindex once you have 5+ articles and intro copy. An indexed category ranking for nothing is a quality liability, not a placeholder.
- Is
noindexthe same as blocking in robots.txt?: No, and mixing them up backfires.robots.txtblocks crawling;noindexblocks indexing. If you block a URL in robots.txt, Google can’t crawl it to see the noindex tag — so the page can still get indexed from inbound links with no description. Use noindex for “don’t show this”, and let Google crawl it.
Related
- Should tag pages be noindex?
- robots.txt — what to put
- noindex vs nofollow explained
- Canonical URLs explained
- Fixing “Crawled — currently not indexed”
- Reading the Pages report
Tags: #Indie dev #SEO #Technical SEO #Indexing #Pillar / Cluster