WordPress Categories vs Tags: How to Structure Your Archives

Almost every WordPress site eventually develops a taxonomy problem. It starts innocently — you publish a post, WordPress asks for a category, you invent one on the spot, then add four tags because the box was there. Repeat that two hundred times and you have thirty unplanned categories, six hundred tags, and a set of archive pages that confuse readers and search engines in roughly equal measure. The fix isn’t complicated, but it does require deciding what categories and tags are actually for.

Below: the real difference between the two, how many of each a site should have, the mistakes that quietly damage SEO, when archives deserve to be indexed, and how to clean up an existing mess without breaking your URLs.

The actual difference

WordPress ships with two built-in taxonomies for posts, and they are not interchangeable variations on “keywords.”

• Categories are hierarchical. They can have parents and children, every post gets one whether you choose or not (WordPress falls back to “Uncategorized”), and they describe the broad section a post belongs to. They’re the table of contents of a book — a short, stable list of the main areas your site covers.
• Tags are non-hierarchical. Optional, flat, and meant to describe specific things a post mentions or concepts that cut across sections. They’re the index at the back of the book — cross-references that let someone who cares about one narrow thing find every place it comes up.

That metaphor answers most day-to-day questions on its own. A table of contents with two hundred entries is useless; so is an index where every entry points to exactly one page. Both only earn their keep when they group things.

On a cooking site, “Desserts,” “Weeknight Dinners,” and “Baking Basics” are categories — every recipe lives in one of those sections. “Chocolate,” “gluten-free,” and “under 30 minutes” are tags, appearing across sections so a reader can slice the site a different way. A gluten-free dessert and a gluten-free dinner sit in different chapters but share an index entry.

Categories vs tags at a glance

How many categories should a site have?

There’s no rule in WordPress itself, but there is a practical answer: few enough that you can recite them from memory. For most blogs and business sites that’s five to twelve. If you can’t remember your own category list, neither can your readers, and it has stopped being a table of contents.

A good test when you’re tempted to add one: will this category realistically hold ten or more posts within the next year? If not, it’s a tag or a subtopic of an existing category. Use child categories sparingly, and only when a parent has genuinely grown too large to browse — three levels deep usually means you’re modelling something that should have been a tag or a custom taxonomy.

Why a post should usually have one category

WordPress lets you tick five categories on a single post. Resist. Filing one post under several does three unhelpful things: the same content appears on multiple archive pages (so those archives increasingly overlap), it muddies breadcrumbs and any category-based URL structure, and it defeats the point of a table of contents — a chapter that contains every page isn’t a chapter.

Pick one primary category per post. If a post genuinely straddles two sections, that usually means either your categories overlap and should be merged, or the second relationship is a cross-reference — precisely what tags are for. Most SEO plugins let you designate a primary category when a post has more than one; set it, because that’s what breadcrumbs and structured data follow.

The classic mistakes

These show up on nearly every site we audit, and they’re all variations on the same misunderstanding.

• A unique tag for every post. A tag that applies to exactly one post isn’t grouping anything — it’s a duplicate of that post with worse content. Sites with 400 posts and 900 tags are the standard version of this.
• Tags that duplicate category names. Filing a post in the “SEO” category and tagging it “SEO” creates two archives listing nearly identical posts. Pick one home for that term.
• Using tags as keywords. Tags aren’t meta keywords and haven’t influenced rankings that way in well over a decade. Fifteen of them on a post does nothing but generate fifteen thin pages.
• Treating both as the same tool. If your categories and tags contain roughly the same words, you don’t have two taxonomies — you have one, entered twice.
• Never deleting anything. A term that made sense in 2019 and hasn’t been used since is clutter now.

What this does to SEO

Taxonomy sprawl isn’t a cosmetic problem. It produces three specific, measurable issues.

Thin archive pages

A tag archive listing one post has essentially no unique content: a heading, an excerpt, a link — text that already exists elsewhere. Publish hundreds and you’ve filled your site with pages that offer a searcher nothing. Quality gets assessed site-wide, not just page by page, so a pile of near-empty pages isn’t neutral.

Index bloat and crawl waste

Every indexable archive is another URL to crawl, and pagination multiplies it — /tag/chocolate/page/2/, /page/3/, and so on. On a big site that burns crawl budget better spent on articles; on a small site it mainly makes Search Console harder to read, with your valuable pages buried among listing URLs.

Cannibalization between an archive and a post

This is the subtle one. Say you have a strong article titled “WordPress Speed Optimization” and a category archive at /category/wordpress-speed-optimization/. Both target roughly the same query, so search engines have to pick — and they may pick the archive, which converts far worse because it’s a list of links rather than an answer. You’ve split your own ranking signals for no benefit. Our guide to running a WordPress content audit covers spotting this pattern site-wide; the taxonomy version is easy to miss because archives don’t feel like “content.”

Noindex or not? The honest answer

The advice you’ll usually see is “noindex all tag archives.” That’s a decent default, but it’s a heuristic, not a law. The better question about any archive is simple: would a searcher be satisfied landing here? Not “is it technically duplicate content,” but whether the page does a job.

• Noindex archives listing one or two posts, ad hoc tag archives, date archives (nobody searches for your March 2023 output), and author archives on single-author sites, where they duplicate the blog index.
• Keep indexed category archives that represent a real section with a decent body of posts — especially with an intro paragraph, a sensible title, and a deliberate order. A “WordPress Security” archive holding twenty solid articles is a legitimate landing page for a broad query, and it can rank.
• Consider keeping the few tag archives that behave like real topic pages. If “gluten-free” covers sixty recipes, that archive serves a searcher better than any single recipe. Treat those as pages: introduce them, curate them, let them compete.

Most SEO plugins expose per-taxonomy indexing controls under a “search appearance” or “indexing” section, and defaults vary by plugin and version — check what yours is actually doing rather than assuming. Our on-page SEO checklist puts these archive-level checks alongside the per-post work.

The rule isn’t “archives are bad.” It’s that an archive should either be worth ranking or be kept out of the index. What hurts is the third option: hundreds of indexed pages nobody curated and nobody would want to land on.

Archives are not a substitute for real internal links

People often justify tag sprawl as internal linking — “the tags connect my posts.” Only weakly. An archive link is one listing among many, with generic anchor text (the tag name), usually in a footer block readers skim past. That’s a far cry from a contextual link inside a paragraph, where the surrounding sentence tells readers and search engines why the two pages relate.

Real links between related posts do the structural work taxonomies get credit for. If keeping those current across a growing archive is the part that keeps slipping, our own free plugin JnK Linkweave (disclosure: we build it) applies contextual linking rules automatically as you publish. It’s the same mechanism behind a well-built topic cluster — categories give a cluster a shared home, but the in-content links hold it together.

Cleaning up an existing mess

If you’ve inherited hundreds of terms, don’t start deleting. Work in this order.

• Inventory first. Open Posts → Categories and Posts → Tags and sort by post count. A handful of terms carry most of your content; a long tail carries one post each. That tail is your cleanup list.
• Merge duplicates. You’ll find “seo,” “SEO,” and “search-engine-optimization” living separately. Merging is safest — posts keep their grouping and the surviving term gets stronger. WordPress has no native merge, but bulk term-management plugins handle it.
• Demote over-specific categories. A category with two or three posts that isn’t a real section of your site is a tag. Reassign its posts, tag them, remove the category.
• Delete dead terms, then redirect. Deleting a term does not delete its posts — it removes the grouping (orphaned posts fall back to your default category) and the archive URL, which starts 404ing. If that URL had traffic or external links, 301 it to the closest surviving archive. If it had neither, a 404 is fine; don’t redirect for the sake of it.
• Curate the survivors. For archives you’re keeping indexed, write a real term description (most themes display it above the list), set a sensible SEO title, and check the page reads like a landing page rather than a database dump.

Permalinks and structure

A couple of structural choices are worth settling early, because changing them later means redirects.

By default, category archives sit at /category/name/ and tags at /tag/name/. Both bases are editable under Settings → Permalinks, and some SEO plugins offer to strip the category base so archives live at /name/. Cleaner, but it raises collision risk with pages and posts at the root — a defensible choice, not a free win, and rarely worth making on an established site for aesthetics alone.

The bigger decision is whether to put the category inside your post URLs (/%category%/%postname%/). It reinforces hierarchy, but hard-couples every post URL to its category: recategorize later and the URL changes, so you need a redirect. It also breaks down for posts in multiple categories. For most sites a flat /%postname%/ is lower-maintenance, with breadcrumbs carrying the hierarchy signal instead. Whatever you choose, change it once and redirect properly — permalink churn costs more than the elegance it buys.

The short version

Categories are your table of contents: few, planned, hierarchical, one per post. Tags are your index: optional, flat, worth creating only when they group several posts. Keep the archives a searcher would be glad to land on, noindex the rest, and prune the taxonomy occasionally the way you’d prune content. Half an hour of decisions now saves an afternoon of cleanup later — and stops your site competing against itself.

Frequently asked questions

What is the difference between categories and tags in WordPress?

Categories are hierarchical and describe the broad section a post belongs to — think of them as your site’s table of contents. Every post gets one, and WordPress assigns “Uncategorized” if you don’t choose. Tags are flat, optional cross-references that describe specific things a post mentions, like an index at the back of a book. Categories group posts into chapters; tags connect related posts across those chapters.

Should I use both categories and tags on the same post?

Yes, but for different jobs. Give each post one primary category that says which section of your site it belongs to, then add tags only for specific themes that genuinely recur across several posts. Avoid tagging a post with the same word as its category — that creates two archive pages listing nearly identical content. If a tag would apply to only one post, skip it.

Should category and tag archives be noindexed?

It depends on whether the archive is useful on its own. Noindex thin archives that list only one or two posts, plus date archives and author archives on single-author sites. Keep well-populated category archives indexed — a curated section archive with a real description and twenty solid posts can rank as a legitimate landing page. The same applies to the few tag archives that behave like genuine topic pages.