What “Doorway Pages” Cost a Behavioral Health Site (And How to Tell If You Have Them)

A treatment center we audited last spring had 47 pages live on the site with titles like “Drug Rehab in Pasadena,” “Drug Rehab in Glendale,” “Drug Rehab in Burbank,” and 44 more. The pages were 320 to 380 words each.

Forty-six of them shared the exact same template, with the city name swapped in seven places and a single line about “serving the [City] community” added at the bottom. The facility itself had one physical address, in Orange County, with no offices, staff, or operations in any of those 47 cities.

That site received a manual action three weeks after our audit started. The Search Console notification cited “Pure spam: Doorway abuse.” Organic admissions inquiries dropped 81 percent in the following 30 days. The fix took four months and a partial rebuild.

The agency that built the 47 pages had pitched them as a “local SEO program.”

If you are a treatment center operator and any part of that story sounds familiar, the next 2,000 words are worth reading carefully. Doorway pages have been a Google policy violation since 2015, but the enforcement reality in 2026 has shifted.

The March 2026 core update wiped out city-page volume across multiple behavioral health sites we monitor. The April 2026 spam-report rule change means a competitor or former employee can now trigger a manual action by reporting your doorway pages directly, with the submission text forwarded verbatim to your site.

The cost-of-getting-caught math has changed, and most operators have not updated their site to match. This is the same pattern Webserv’s SEO program flags first when we audit a new behavioral health site.

Below is what doorway pages actually are, how they show up in this category, how to tell if you have them, and how to fix them without losing the rankings you do have.

What Google actually means by “doorway pages”

The official definition lives in Google’s spam policies documentation under the heading “Doorway abuse.” The policy describes the pattern across four signals.

The signals are: multiple sites or pages with slight variations targeting the same query intent, and multiple domains funneling users to one destination. Also flagged: pages generated to funnel users into the actual usable section of a site, and near-identical pages that sit closer to search results than to a clearly defined site hierarchy.

The intent test is what matters. Google is not banning location pages. It is banning location pages built primarily to rank, not to serve the user once they land.

The line the policy draws is between content that exists to answer a query and content that exists to capture a query.

A real location page describes a real facility at a real address with real staff serving real patients. It carries unique local information that could not have been generated by template. A real local SEO playbook starts here, not at programmatic page generation.

The same logic applies to the rest of the commercial layer. A real service page for PHP, IOP, or residential treatment is built around the specific clinical reality of that program, not around a keyword combination. A doorway page describes a generic offering with a city name swapped in. The page exists because the operator wanted to rank for “[service] in [city],” not because the operator has anything specific to say about that city.

Most operators do not set out to build doorway pages. They hire an agency. The agency proposes a “local SEO” program. The program produces 30 to 60 city pages on a template. The operator approves them because they look like SEO work.

Google’s quality systems flag them because the pattern they form is doorway abuse.

How doorway patterns show up on behavioral health sites

Five patterns account for most of the doorway issues we see in this category.

Templated city pages with no local content. A facility in Costa Mesa publishes “Drug Rehab in Anaheim,” “Drug Rehab in Long Beach,” “Drug Rehab in Riverside,” and 40 more cities the facility does not physically operate in. Each page describes the facility’s program with the city name dropped in. There is no local address, no local staff, no local payer information, no local context.

Insurance + service combinations on a template. Pages like “Aetna Drug Rehab in [City],” “BCBS Detox in [City],” and “Cigna Outpatient in [City]” multiplied across a payer matrix. The pages exist to capture insurance + service + geo queries. The content is the same boilerplate insurance explanation with the variables swapped in.

Level-of-care + city combinations. “PHP in [City],” “IOP in [City],” “Residential in [City]” multiplied across 30 cities. Same template, same content, only the level of care and city change.

Modality + city combinations. “TMS Therapy in [City],” “Ketamine-Assisted Therapy in [City],” “EMDR in [City]” multiplied across cities. Often these are even worse because the underlying modality content is itself a thin description with no clinical depth.

Microsite networks. A single operator running 4 to 8 separate domains, each targeting a different city or region, all funneling calls to the same intake line. This is the heaviest pattern Google’s policy specifically names. It is also the easiest pattern to detect from outside, and the most likely to trigger a competitor spam report.

The pattern shows up most often on sites built by legacy SEO agencies in this category that productized programmatic page generation a decade ago and never updated the playbook.

If your site does any of the first four at scale, the doorway risk is significant. If your operation runs the fifth, the risk is severe and the timeline for action should be measured in weeks, not quarters.

How to tell if you have doorway pages

The self-audit is short. It takes about an hour and does not require any tooling beyond a browser and a spreadsheet.

Step 1: List every location, insurance, and modality page. Pull a site:yourdomain.com query in Google or export the URL list from your sitemap. Filter for any URL containing a city name, payer name, or modality name. This is your candidate set.

Step 2: Open five of them side by side. Pick five at random from the candidate set. Read them top to bottom. Note what content is unique to each page versus what is shared boilerplate.

Step 3: Run the 30-percent test. For each pair of pages, estimate the percentage of body content that is unique to one page versus shared with the other. If the unique content is under 30 percent across the pairs, you have doorway pages.

If most pairs land between 30 and 60 percent, you have a thin-content problem that is one Google update away from being treated as doorway abuse. Above 60 percent is the safe zone.

Step 4: Check the H1, meta title, and meta description. If the only difference between pages is the city, the payer, or the modality variable, the page was built to rank, not to serve. That is the doorway signature.

Step 5: Check for a real local signal. A genuine location page carries a local address, local staff bios, local payer mix details, local trust signals, and local geographic context like neighborhoods or counties served. If none of those appear, the page is not a location page in any meaningful sense. It is a doorway with a city name on the door.

A useful sanity check: ask whether a human visiting the page would learn anything they could not learn from the homepage by adding the city to a search. If the answer is no, Google’s quality systems are likely to reach the same conclusion.

What doorway pages actually cost you

The cost is rarely a single number. It is layered across four categories, and the total is often two to three years of compounding decline before the operator realizes what is happening.

Direct algorithmic suppression. Google’s quality systems demote doorway-heavy sites during core updates without filing a manual action. The site does not get a Search Console notification. It just stops ranking. The March 2026 core update wiped out doorway-heavy city-page volume across multiple behavioral health sites we monitor. Operators on those sites saw 40 to 70 percent organic traffic drops with no notification and no obvious cause.

Manual action risk. If the pattern is egregious or if the site has been reported, Google can issue a manual action under “Pure spam: Doorway abuse.” The action affects the entire site, not just the offending pages. Per Google’s Search Console Help documentation on manual actions, the recovery process requires removing or fundamentally rewriting the doorway content before submitting a reconsideration request.

Cannibalization of legitimate pages. If you have one real, well-built location page for the city where your facility actually operates, and 30 templated city pages alongside it, the templated pages dilute the equity that should consolidate on the real one. Google sees 31 weak signals for the city instead of one strong one. The legitimate page often ranks worse because of the templated company it keeps. This is the same dilution problem topical authority frameworks try to solve at the cluster level.

Crawl budget waste. Every templated city page consumes crawl budget that should be spent on pages that earn admissions. On a site with 60 doorway pages and 12 real ones, Google’s crawler is spending most of its time on content that hurts the site.

The real pages get reindexed less often, updates take longer to register, and the site’s overall responsiveness in search degrades. The problem compounds on sites with mobile-first indexing issues already in play, because the doorway pages also waste mobile crawl cycles.

The compounding cost across these four is the part most operators underestimate. A single year of algorithmic suppression on a behavioral health site with $40,000 per admission economics is six to seven figures of admissions revenue that never showed up.

How to fix doorway pages without losing the rankings you do have

The wrong move is to delete every templated page and watch what happens. The doorway pages have probably accumulated some links, some traffic, and some rankings. Deleting them without a plan throws away the equity along with the problem.

This is also why a full site rebuild rarely helps for sites with doorway problems: the consolidation pass below preserves more equity than a rebuild would.

The right approach is a four-step consolidation pass.

Step 1: Identify the one or two pages per cluster worth saving. For the city-page cluster, the page worth saving is usually the city where your facility actually operates. For the insurance-page cluster, it is the page for the payer that drives the most admissions. For the modality cluster, it is the modality your clinical team is most known for. Everything else gets consolidated into those saved pages.

Step 2: Rebuild the saved pages with real local, payer, or clinical content. A real city page includes the physical address, photos of the actual facility, staff bios for the people working at that location, payer details specific to that market, and geographic context (counties served, surrounding cities, public transit notes, nearby hospitals or clinical partners). The page should be 1,200 to 2,000 words of content that could not have been generated by template.

Step 3: 301 redirect the doorway pages to the rebuilt versions. Every templated city page gets a 301 to the rebuilt page for that cluster. This preserves the link equity from any incoming references and signals to Google that the consolidation is intentional.

Do not let the doorway pages go to 404 or to the homepage. The redirect target should be the closest topical match in the consolidated set.

Step 4: Update internal links and sitemap. The internal links pointing to the deleted doorway pages need to be updated to the rebuilt destinations. The XML sitemap needs to be regenerated without the deleted URLs. The technical SEO pass that catches the orphan internal links should run on the same day as the redirects to keep the site clean.

This sequence usually recovers 60 to 80 percent of the pre-decline organic visibility within 90 to 120 days. The remaining 20 to 40 percent depends on the site’s broader technical SEO foundation and on whether the operator can resist the agency pitch to rebuild the doorway pattern under a different name a year later.

The line between programmatic location pages and doorway pages

This is the question most operators ask once they see the pattern: where is the line between a programmatic page strategy that works and one that gets penalized?

The line is differentiation, not volume. Google does not have a policy against scale. It has a policy against pages that exist to rank without serving.

A site can run 200 location pages legitimately if each page carries enough unique, local, useful content that a human visiting it would learn something specific to that location. The same site can run 10 location pages illegitimately if each one is a city-name swap on a template.

The clearest line is the address test. If you have a real facility at the location, you can build a real page for it.

If you do not have a real facility at the location, the page has to either be honest about that (a “service-area” page that says clearly “we serve this region from our [actual city] facility”) or it should not exist. Pretending to have a facility where none exists is also a healthcare advertising compliance problem, not just an SEO one.

A second test is the staff test. A real location page can name actual people who work at that location, link to their credentials, and show their photos. A templated city page cannot do that because the staff do not exist there.

Staff bios are the cheapest, fastest way to differentiate a location page from a doorway page, and per Google’s people-first content guidance, quality systems specifically weigh staff and authorship signals on YMYL pages.

The third test is the payer test. A real location page can list specific in-network payers for that market, with payer logos and verification information. A templated city page cannot because the payer relationships are tied to the facility’s actual contracts, not to the city.

If your location pages cannot pass all three tests (address, staff, payer), the program is doorway-adjacent and the safer path is consolidation, not expansion. The same principle is why we recommend service-pages-first ordering on every audit: one strong commercial page beats many thin ones, every time.

The pull quote is uncomfortable because the math is uncomfortable. The cost of getting caught in 2026 is higher than the cost of having fewer pages, and most operators have not done the comparison. The broader behavioral health marketing program only compounds when the foundation is clean. Doorway clusters are the most common reason it does not.

Frequently Asked Questions

Trevor Gage is the Director of Earned & Owned Media at Webserv, where he leads SEO, content strategy, and organic acquisition for behavioral health treatment centers. He has overseen audits of more than 200 behavioral health sites and writes about the technical and editorial SEO standards Webserv applies to client work.