Doorway page
A doorway page is a page made to rank for a narrow variant, then funnel visitors to the same place. Think dozens of near-duplicates for each town, feature, or misspelling that all point to one signup. For a small site, this risks Google’s spam systems and wastes crawl and effort that could go to real pages.
By Théophile Louvart, founder of Porteur · Updated 14 September 2026 · Markdown
What counts as a doorway page
Google’s spam policies define doorway pages as many pages that funnel users to one destination. They exist to manipulate rankings, not to help people.
Patterns that trigger it: near-duplicate “city” or “keyword variant” pages where only a token changes, doorway hubs that list the variants then all point to one signup, or cookie‑cutter branches of a site that lead to the same content after a click.
A real page family is different. Each page stands on its own with unique answers, data, or offers that a reader would want by itself. A page earns its place when it holds substance, not just a swapped noun.
City and variant patterns that cross the line
- /locations/london, /locations/leeds, /locations/luton where the only change is the city name, the body is the same, and all CTAs lead to /signup
- /features/crm-for-fitness, /features/crm-for-salons, /features/crm-for-plumbers that all repeat the same copy and link to /pricing
- /guides/how-to-use-yourproduct-in-{city} that give no local info, just a thin intro and a link to a generic guide
A safe variant shows clear, page‑level value: unique local data, service coverage, case studies, pricing or hours that differ, legal notes, or inventory that changes per city. Example: /locations/london has real areas served, travel times, local testimonials, and a booking calendar for London only, while /locations/leeds has Leeds‑specific content and availability.
How to spot doorway risk on your site
List suspect folders
Export URLs under /locations, /areas, /features-for-… or any templated path. A sitemap or a crawl works.
Check uniqueness
Open five random pages. If titles, H1s and paragraphs are near-identical with only a token swap, mark them.
Trace the funnel
If most CTAs on those pages all send to the same /signup or /pricing with no page‑specific conversion path, add to the risk list.
Review Search Console
Look for many pages with impressions but no clicks, or clusters that dropped after a spam update. Check Manual actions for any notice.
Assess intent overlap
Search the variant queries in quotes. If the user need is the same across variants and your pages add no distinct value, they are likely doorway pages.
Fixes that keep useful pages and remove spam patterns
- Merge thin variants into one page that honestly covers the topic, then 301 redirect old URLs to the best page
- Rebuild high‑value variants with unique local or segment data, proof, and offers, not just a swapped word
- Consolidate internal links: link to the canonical page, not to many near-duplicates
- Remove doorway hubs that only link to variants without offering value
- Update templates so each new page demands page‑level inputs: local stats, inventory, testimonials, distinct FAQs
A fixed page looks like /features/crm-for-fitness with sector metrics, named workflows, a case study, and feature emphasis that differs from /features/crm-for-salons. Each has its own CTA to a relevant demo, not a generic signup.
How it differs from a real page family
| Aspect | Doorway pattern | Real page family |
|---|---|---|
| User need | Same need across pages | Distinct needs per page |
| Content | Token swaps and boilerplate | Unique data, answers, offers |
| Destination | All funnel to one generic page | Each converts on-page or to a matching flow |
| Discovery | Many pages with impressions, few clicks | Fewer pages, stronger clicks and engagement |
For a small site, fewer strong pages beat many weak ones. This also reduces index bloat and makes internal linking cleaner.
Questions
They are many near-duplicate pages made to rank for slight variants, then funnel visitors to the same destination. Google treats them as spam. They waste crawl and can trigger a ranking drop or a manual action.
A landing page solves a specific need and can stand alone. A doorway page is thin, templated, and exists to capture traffic for a token change, then sends everyone to the same place. If each page would still be useful without the others, it is likely a real landing page.
Crawl templated folders and scan for near-identical titles and copy with only a city or segment swapped. In Search Console, look for clusters with impressions but few clicks, and any manual action. Check where the CTAs point. If they all go to /signup or /pricing, review those pages.
Clean-ups can help, but recovery often waits for a later core or spam update. If you had a manual action, you must fix the problem and submit a reconsideration request. Keep improving the remaining pages so they offer clear, page‑level value.
Yes, if each page has real local value: accurate service areas, different availability, local case studies, local regulations, or stock that changes by city. If you cannot provide that, publish one strong page that explains coverage and how you serve each area.
Disavow is rarely needed. Google says it already ignores most bad links and it is meant for sites with a link spam manual action or a history of paid links they cannot remove. Focus on fixing content and structure first.
Sources
Check my site, free
Not sure if your city or feature variants cross the line, run your URL and see three doorway risks or opportunities in about thirty seconds, free.
- Free check, no card
- Read-only, your own accounts
- Readable by your agent
Read next
- GuideGoogle spam updates: the policies they enforce and how to stay clear
- GuideManual actions and the reconsideration request
- GuideProgrammatic SEO for a small product: page families that earn
- GuideKeyword cannibalization: two pages, one search, and how to fix it
- GlossaryIndex bloat
- GlossarySemantic search
- GlossaryListicle
- GuidePricing page examples: the patterns that convert, and why they work