Duplicate page titles

A template constant. Every page gets the site name because the title field was never wired to the page's own data. Easy to spot: the duplicate count equals the page count of that template.

Paginated or filtered listings. Page 2, page 3 and every colour filter inherit the category title. Common, and usually harmless if canonical tags are right — but the titles still make the results unusable if those URLs are indexed.

Genuinely duplicated pages. The same content reachable at several URLs — with and without a trailing slash, with tracking parameters, under two categories. Here the title is telling you about a URL problem, not a title problem.

For a searcher, several identical results are noise: they cannot tell which one answers their question, and the site looks careless.

For a search engine, identical titles on similar pages contribute to a judgement that the pages are duplicates, which can lead to only one being indexed. Google documents that it rewrites titles it finds unhelpful, and repetition across a site is one of the cases it names.

The diagnostic value is the real point: a sudden block of duplicate titles after a release is one of the fastest ways to notice that a template shipped broken.

Collect titles for every URL — from a crawl or your CMS export — and count distinct values. The ratio of distinct titles to pages is the number that matters; anything far below one deserves a look.

Then group the duplicates and ask, for each group: do these pages have the same content? If yes, the problem is duplication, not titling. If no, the template is at fault.

Template fault: include a field that varies — the product name, the article title, the city — and make sure the fallback also varies.

Pagination: append the page number, and make sure the paginated URLs are either indexable with distinct titles or excluded deliberately. Both are defensible; silence is not.

Real duplicates: pick one canonical URL and point the others at it. Fixing the titles alone would hide the symptom and leave the duplication.

`seo.duplicate_titles` compares the titles of every page the crawler reached and reports groups sharing one, with the URLs in each group. The grouping is what makes it usable: it shows the shape of the problem — one template, one listing, or scattered pairs.

The finding sits next to `seo.canonical`, because the two answer the second question: whether the duplicated pages already declare a canonical URL.

Adding a distinguishing field is the easy half. Choosing one that actually distinguishes is the half that gets skipped.

A location directory that titles every page `Plumber in {city} — Acme` has unique titles and no useful variation: the pages are identical apart from one word, and a search engine judging them near-duplicates is not wrong. The fix there is not a better title formula, it is content that differs per city.

The reverse mistake is loading the title with every attribute available. `Blue running shoes size 42 mens lightweight breathable — Acme Sports Equipment Store` is unique, unreadable, and long enough to be truncated before the brand appears.

The rule that survives contact with real catalogues: put the single most distinguishing attribute first, add at most one qualifier, and let the rest of the page carry the detail.

Order matters more than length because of truncation. Everything after roughly sixty characters is invisible in most results, so a brand prefix spends the most valuable part of the title on the word that is identical across the whole site.

And where two pages genuinely resist being told apart in a title, that is worth treating as information rather than as a formatting problem: it usually means they should be one page.