What Is Thin Content and How Do You Find Yours?

Jul 19, 2026 · SEO Glossary

Here's a pattern I keep finding in audits of local-service sites: forty near-identical pages titled "Plumber in Austin", "Plumber in Round Rock", "Plumber in Cedar Park" — same 150 words, city name swapped by template. The owner paid an agency good money for those pages and can't understand why traffic fell after they went live. Those pages are the textbook answer to what is thin content, and Google has been algorithmically demoting exactly this pattern since the Panda update in 2011 — logic that's long since been folded into its core ranking systems.

What Is Thin Content in Google's Eyes?

Thin content is a page that adds little or no value for the person who lands on it. Note what the definition is not: it's not about length. A 3,000-word page assembled from paraphrased competitor articles is thin; a 120-word page that states a shipping policy clearly is not. Google's own documentation on manual actions calls out "thin content with little or no added value" — the operative phrase being added value. The question a quality rater asks isn't "how many words?" but "did this page give the visitor anything they couldn't get from ten other results?"

The Four Classic Varieties

• Doorway pages. The city-swap pattern above — sets of pages each targeting a keyword variation, all funneling to the same offer, differing only in the substituted term. Google treats these as spam, and they're explicitly named in its policies.
• Scraped or lightly rewritten content. Republishing others' text, spun paraphrases, or — the 2026 version — unedited AI output that restates what the top ten results already say. The generation method isn't the offense; adding nothing is.
• Auto-generated filler. Tag archives with one post, empty category pages, search-result pages indexed as content, boilerplate location pages generated from a database.
• Thin affiliate pages. Product pages that reproduce the merchant's description, add an affiliate link, and contribute no original testing, comparison, or opinion.

Why a Few Bad Pages Poison the Whole Domain

Google evaluates quality at the page level and in aggregate. The helpful content system, rolled into core updates since 2024, applies a site-wide classifier: a site with a high proportion of unhelpful pages sees its helpful pages ranked lower too. Think of it as a trust ratio — 300 thin tag archives sitting next to 40 good articles means most of what Google samples from your domain is junk, and it scores your domain accordingly. This is also where E-E-A-T enters: pages demonstrating first-hand experience are the opposite of thin, and sites accumulating them build exactly the signals the quality systems reward. Deleting or fixing thin pages is one of the few SEO actions that can lift rankings on pages you didn't touch.

Finding It: A Repeatable Method

You can't fix what you haven't inventoried, and eyeballing the sitemap doesn't scale. The workflow I use:

1. Crawl everything. Run the site through a site auditing tool to get a complete URL inventory with word counts, titles, and duplicate-content groupings. Sort ascending by word count and flag near-duplicate title patterns — the city-swap template lights up immediately.
2. Layer on engagement data. Export twelve months of landing-page data from GA4 and Search Console. Pages with near-zero impressions and near-zero entrances for a year are contributing nothing except crawl load.
3. Check what Google indexed. The Pages report in Search Console shows how many URLs sit in "Crawled — currently not indexed". A big number there is Google already telling you it considers those pages not worth keeping.
4. Judge the survivors by hand. Metrics shortlist candidates; a human decides. A low-traffic page answering a rare but real question is fine. A high-traffic page that's embarrassingly shallow is still a liability.

Improve, Merge, or Remove — the Decision Rule

For each flagged page, ask: does a real query deserve this page as its best answer?

• Yes, and it could be that answer → rewrite it properly. Add the original substance it lacks: your own data, photos, process, pricing, opinion.
• Yes, but a sibling page already covers it → merge. Fold any unique fragments into the stronger page and 301-redirect the thin one to it. Ten overlapping weak posts routinely consolidate into one page that outranks all ten.
• No real query needs it → remove. Delete and return 404 or 410, or noindex pages that must exist for users (internal search results, faceted filters) but shouldn't be in the index.

Resist the urge to bulk-delete everything old, though — age isn't thinness, and pages with backlinks or steady long-tail traffic deserve a rewrite, not a redirect. If you're preparing this analysis for a client, our SEO audit report sample shows how to present the inventory and the improve/merge/remove call for each cluster without burying them in spreadsheets.

Keeping It From Growing Back

Thin content is usually a system output, not an authoring mistake — a CMS auto-creating a tag archive per tag, a template generating a page per database row. Fix the generator: noindex thin archive types by default, require a minimum of genuine content before a location page goes live, and add word-count and duplicate-title checks to your recurring crawl so regrowth shows up in a report instead of in a ranking drop.