Thin Tag Archives and the WordPress Pages Nobody Meant to Publish

A partner asks why the eight practice area pages the firm paid for last spring still are not pulling traffic. The content is fine. The writing is better than most of what the competition has. The pages are indexed. And yet the site behaves like a place Google visits out of politeness rather than interest.

Then you crawl it and find that the firm’s WordPress install is publishing 640 URLs, of which roughly 90 are pages anyone at the firm actually intended to exist. The rest are author archives, monthly date archives, attachment pages, paginated tag listings, and forty tags that each contain exactly one post. Nobody built those. WordPress built them, quietly, as a byproduct of normal publishing, and the firm has been feeding them to search engines for six years.

This is one of the most common conditions we find on established law firm websites, and it is one of the few technical problems where the cleanup is cheap, reversible, and usually a precondition for the content work paying off.

Where All Those URLs Come From

WordPress is a publishing system with a generous default. Every time you create a piece of content, it also creates several ways to list that content. Those listing pages are real URLs, they return a 200 status, and unless somebody told them otherwise they are open to indexing.

The four usual sources on a firm site look like this.

• Author archives. Every user account with a published post gets a URL at /author/name/. On a firm where the marketing coordinator, an old agency login, and three attorneys have all published something, that is five archives. Two of them list a single post. One is named after a vendor the firm stopped working with in 2021, which is now a public page on the firm’s domain.
• Date archives. WordPress generates yearly, monthly, and often daily archives. A blog running since 2018 with occasional posting produces roughly 90 monthly archive URLs, most holding one or two posts, plus year archives on top. None of them answer a question a prospective client has ever asked.
• Attachment pages. This is the one that surprises people. Every image uploaded to the media library can get its own URL, a bare page containing the image and nothing else. A firm with 400 uploaded images can be sitting on 400 indexable near-empty pages. Newer WordPress versions redirect these to the file itself by default, but sites carried forward through years of upgrades and page builders very often still serve them.
• Tags applied once and never again. Somebody tagged a post “car accident” and another “car accidents” and another “auto collision.” Each spawns an archive listing one post. Multiply by five years of inconsistent tagging and forty single-post tag archives is a conservative number. Tag archives also paginate, so a handful of them spawn /page/2/ variants too.

Add feed URLs, search result pages that got linked from somewhere, and the paginated blog index, and a firm with 90 pages of real content is asking search engines to evaluate seven times that volume.

Why This Actually Costs You Something

Be careful with the mechanism here, because this topic attracts a lot of confident nonsense. Thin archive pages are not a penalty. There is no algorithmic hammer that drops on a site for having date archives. What actually happens is duller and more expensive.

First, crawl allocation. Google decides how much attention a domain is worth and spends it. A firm site is not a news publisher, so the crawl budget is modest. When most of the URLs on offer are auto-generated lists, a meaningful share of that attention goes to pages that will never rank and never convert. In server logs on sites in this condition, we regularly see 40 to 60 percent of Googlebot requests landing on archive and attachment URLs. The new practice area page that went live last Tuesday gets discovered slower and revisited less.

Second, quality signals in aggregate. Search engines form a view of a site as a whole. A domain where the majority of indexable URLs are near-empty is a domain where the average page is weak, and that shapes how readily anything new from that domain is trusted. This is not a scored penalty, it is an impression, and impressions are harder to argue with.

Third, and most concretely, duplication and cannibalization. A tag archive titled “Truck Accidents” containing one post about truck accidents competes with the truck accident practice page for the same query. Sometimes the archive wins, and the prospect lands on a stub listing one blog post instead of the page built to convert them. We have seen firms lose form fills for months to a tag archive that outranked the service page it was drawn from.

Fourth, diagnostics get useless. Search Console’s Pages report fills with hundreds of “Crawled, currently not indexed” and “Duplicate without user-selected canonical” entries, all archives, and the two real pages with an actual problem are buried in the noise. Firms in that state usually cannot tell which indexing issues matter, which is a separate and worse problem covered in our piece on when Google will not index your law firm pages.

How to Find Every One of Them

Do not guess. Inventory. This takes about an hour on a typical firm site and the output is a spreadsheet you can make decisions from.

1. Crawl the site. Screaming Frog SEO Spider is the standard tool and the free tier covers 500 URLs, which is enough for a small firm site and will at minimum show you the shape of the problem. Crawl with JavaScript rendering on, then group by URL pattern. Look for /author/, /tag/, /category/, /20xx/, and anything with an image filename in the path.
2. Pull the indexed set from Search Console. Under Pages, export both the indexed and the excluded lists. The excluded list is where attachment pages and date archives usually pile up. Filter the export by the same patterns.
3. Run site queries as a sanity check. A site search restricted to /tag/ or /author/ on your domain gives a quick read on what is actually surfacing. It is imprecise, but it is fast and it tells you whether anything embarrassing is public.
4. Check your SEO plugin’s current settings. Yoast, Rank Math, and SEOPress all have archive controls, and on most sites we inherit, somebody toggled two of the four and stopped. Screenshot the settings before you change anything.
5. Pull the tag list with post counts. In the WordPress admin under Posts and Tags, sort by count. Everything with a count of one or two is a candidate for deletion. Print that list. It is usually longer than the marketing lead expects.
6. Check server logs if you can get them. Thirty days of log data tells you where crawl attention is actually going, which turns this from a theory into a number you can show a partner.

Do all of this before you touch a setting. The point of the inventory is that a few of these URLs have real value and you need to know which ones before the bulk work starts.

What to Noindex and What to Delete

These are different actions with different consequences, and conflating them is how cleanups go wrong.

Noindex and Keep

Noindex tells search engines not to list a page while the page still exists and still works for humans and internal navigation. This is the right call for anything with a legitimate use that simply does not belong in search results.

• Date archives. Noindex all of them. If the firm truly does not use them, disabling them entirely is cleaner, but noindex is the safe default.
• Author archives on a single-author site. If one account publishes everything, the author archive duplicates the blog index. Noindex it.
• Paginated archive pages beyond page one. Keep them crawlable so posts stay discoverable, but out of the index.
• Internal search results pages. Always noindex. These are an old and reliable source of junk URLs.
• Tag archives that hold real depth, meaning eight or more genuinely related posts, are a judgment call. If the tag functions as a topic hub a reader would use, keep it indexable and give it an intro paragraph so it is more than a list of links. Otherwise, noindex.

Delete or Redirect

Deletion is for URLs that should not exist at all.

• Single-post tags. Delete the tag. Deleting a tag does not delete the post, it only removes the association, and the post stays exactly where it is. This is the single highest-volume win available and it is genuinely low risk. If a deleted tag archive had inbound links or measurable traffic, and a small number will, 301 it to the most relevant practice page.
• Attachment pages. Redirect them to the media file or to the parent post. Every major SEO plugin has a one-toggle setting for this. It is the highest ratio of URLs eliminated to effort spent on the entire list.
• Author archives for departed staff and old vendor accounts. Reassign the posts to a current author, then remove the stale account. Leaving a former agency’s name published on a law firm domain is a bad look independent of any SEO consideration.
• Duplicate tags. Merge “car accident” into “car accidents” using a term merge, keep one, 301 the other.

One firm rule before anything gets removed. Check inbound links and check the last twelve months of organic entrances for each URL you plan to delete. Ninety-eight percent of the time the answer is zero and you proceed. The two percent where it is not zero is exactly why you check.

Getting the Settings Right in WordPress

Most of the noindex work is four toggles in an SEO plugin under its search appearance or titles settings, covering author archives, date archives, media pages, and tag archives. Set them, then verify by loading three affected URLs and viewing the source for the robots meta tag. Plugin settings and rendered output disagree more often than anyone likes.

Two structural points worth understanding while you are in there. WordPress’s permalink and taxonomy behavior, including how tag and category bases produce URLs, is documented in the official WordPress permalinks documentation, and it is worth ten minutes of a marketing lead’s time so future decisions are made deliberately rather than inherited.

The second point is that this cleanup is not a substitute for structure, it clears the ground for one. A firm site should route authority through a deliberate hierarchy from practice area hubs down to supporting content, which is the case we make in why law firm websites need silo architecture. Auto-generated archives are the opposite of that, an accidental parallel structure that splits signals across URLs nobody designed. Removing them makes the intentional structure legible. The broader checklist this sits inside is covered in our guide to technical SEO for law firm websites.

What to Expect Afterward, Honestly

Here is the part most write-ups skip. Cleaning up 400 thin URLs does not by itself make a firm rank. It removes a drag and it removes noise. Those are real but modest gains, and anyone telling you otherwise is selling something.

What we typically observe on firm sites over the eight to twelve weeks following a cleanup is faster indexing of new pages, often days instead of weeks, a Search Console Pages report that becomes legible enough to diagnose actual problems, and the disappearance of tag archives from queries where a service page should have been showing. Total organic sessions frequently dip slightly first, because those archives were catching a trickle of low-intent traffic that never converted. That dip is expected. Set the expectation with the partner before you start, not after the first monthly report.

The real return shows up in the next thing you publish. When a firm invests in eight substantial practice area pages, those pages need to be crawled promptly, evaluated against a domain that looks curated rather than cluttered, and left to compete without a stub archive undercutting them. That is why we run this cleanup before content work rather than alongside it. The order matters, and doing it in the other order is how firms conclude that content does not work for them.

Budget half a day to a full day of technical work for a typical firm site, plus the hour of inventory. There is no ongoing cost. Then set one recurring habit, a quarterly look at the tag list with post counts, deleting anything that landed at one, so the site does not rebuild the same problem over the next three years.

Find Out What Your Site Is Actually Publishing

Most firms have never seen a full URL inventory of their own website, and the number is almost always three to seven times higher than the marketing lead’s estimate. That gap is not a crisis, but it is worth knowing about before you approve another content budget, because it changes what that budget buys. Rubiks crawls the whole property, separates the pages you meant to publish from the ones WordPress published for you, and tells you which ones to noindex, which to delete, and which few are worth keeping. Talk to Rubiks and we will show you your real page count before you spend another dollar on content.