Restore / Research / Cloaking SEO Cloaking Reference Cloaking SEO. How Google detects conditional DOM serving and the remediation path.
Detection runs at crawl time as DOM-parity comparison between Googlebot and headless-browser sessions. Five variant manual-action categories route to their own reconsideration templates. Historical precedent goes back to the 2006 BMW Germany and Ricoh Germany de-indexing.
Four facts about cloaking detection that govern the remediation.
Cloaking flags surface as manual actions in the Search Console Manual Actions report. The remediation path is full removal of the conditional serving logic, then reconsideration with per-URL DOM-parity evidence. <a class="prose-link" href="/">SEO penalty removal services</a> for cloaking-class actions route here.
- 01
Cloaking is conditional DOM serving by request fingerprint.
A server serves a different HTML Document Object Model to Googlebot than to a human user. The condition is usually a check on the requester's IP address against a database of known Googlebot IPs, or an evaluation of the <span class="cite">User-Agent</span> HTTP header. The intent is to surface high-ranking content to the crawler and unrelated content to the user. Google's policy treats this as a violation regardless of the technical implementation.
- 02
Detection is DOM-parity comparison.
Google compares the rendered DOM of the Googlebot session against automated headless-browser sessions simulating standard user agents. A mismatch beyond reasonable rendering variation triggers an automated flag and often a manual review. The detection runs at crawl time and does not require the operator's cooperation.
- 03
Manual action categories vary by surface.
<span class="cite">Cloaking</span> covers standard HTML DOM mismatch. <span class="cite">Sneaky redirects</span> covers conditional redirects that route Googlebot to one URL and users to another. <span class="cite">Cloaked images</span> covers obscured images or mismatched thumbnails served to Google Images. <span class="cite">AMP content mismatch</span> covers AMP pages lacking topical or functional parity with their canonical web page. <span class="cite">Structured data manipulation</span> covers markup of invisible text or irrelevant entities. Each category routes to its own reconsideration template.
- 04
Historical enforcement set the severity precedent.
In <span class="cite">February 2006</span> Google applied manual actions removing BMW Germany and Ricoh Germany completely from the index for deceptive cloaking and doorway pages. Both domains followed the standard reconsideration workflow (remove the conditional serving logic, submit the reconsideration request) and were restored to the index. The precedent stands: cloaking violations can result in total de-indexing, and the remediation path is reconsideration after full removal of the deceptive logic.
Four phases. Detection in. Reconsideration-ready evidence package out.
-
Run the DOM-parity test.
We fetch the URL as Googlebot and as a standard browser user agent, then compare the rendered DOM. Mismatches beyond reasonable rendering variation surface the conditional serving logic. The audit also reads server logs for evidence of User-Agent or IP-based routing.
- Fetch as Googlebot user agent
- Fetch as standard browser user agent
- Render both and compare DOM at element level
- Review server logs for conditional routing
- Identify CDN edge rules if present
-
Strip the conditional serving logic completely.
Remediation requires removing the conditional serving at every layer: application server, CDN edge worker, reverse proxy, redirect logic. Partial removal is the most common rejection cause. The remediation must produce DOM parity end to end.
- Remove server-side User-Agent branches
- Remove CDN edge-worker conditional logic
- Remove conditional redirect rules
- Audit AMP-to-canonical parity (if AMP shipped)
- Re-verify DOM parity across all affected URLs
-
Author the reconsideration request to the three-element structure.
The reconsideration request names the exact quality issue (cloaking, the specific variant), documents the steps taken to remove the conditional serving logic, and demonstrates outcome documentation with per-URL DOM-parity evidence. The pages must be fully accessible to Googlebot during review with no noindex or paywall.
- Name the manual-action category exactly
- Document the conditional-logic removal across URLs
- Attach per-URL DOM-parity evidence
- Verify full Googlebot accessibility
- Submit through the Search Console interface
-
Verify the action is lifted on the recrawl.
The reconsideration cycle runs on Google's review timeline. We monitor the Manual Actions report for the action's removal and verify on recrawl that the parity logic stays consistent. Recurrence triggers re-flagging on a tighter review window than the original case.
- Monitor Manual Actions report daily
- Track recrawl coverage of affected URLs
- Verify parity holds at next crawl cycle
- Lock the parity test into the build pipeline
Detection mechanics versus remediation requirements on a cloaking-class action.
- Trigger surface
- Automated DOM-parity comparison between Googlebot and headless-browser sessions at crawl time.
- Manual action category
- Routes to one of cloaking / sneaky-redirects / cloaked-images / AMP-content-mismatch / structured-data-manipulation per the surface flagged.
- User-Agent or IP fingerprinting
- Detection ignores the fingerprinting method itself; only the DOM mismatch triggers the flag.
- AMP-canonical parity
- AMP pages lacking topical or functional parity with the canonical web page trigger AMP-content-mismatch and AMP drops from Search.
- Historical precedent
- February 2006 BMW Germany and Ricoh Germany de-indexed entirely. Severity ceiling is total removal from the index.
- Trigger surface
- Complete removal of conditional serving logic at every layer: app server, CDN edge, redirects.
- Manual action category
- Reconsideration request authored to the matched category template with per-URL evidence.
- User-Agent or IP fingerprinting
- Remediation strips every fingerprint-conditional branch even if some branches were not flagged directly.
- AMP-canonical parity
- Restoration requires content parity between AMP and canonical, then reconsideration request, then recrawl confirmation.
- Historical precedent
- Both domains restored after reconsideration. The path exists; the remediation cost scales with the deceptive logic's footprint.
Specification grounded against Google Search Central cloaking policy and the historical 2006 BMW Germany / Ricoh Germany remediation precedent.
What operators ask before scoping the remediation.
- 01.
How does Google detect cloaking?
Google compares the rendered DOM of the Googlebot session against automated headless-browser sessions simulating standard user agents. A mismatch beyond reasonable rendering variation triggers an automated flag and often a manual review. Detection runs continuously at crawl time and does not require operator cooperation.
- 02.
Is showing different content based on geography cloaking?
Geography-based content variation that applies equally to Googlebot and to human users from the same region is not cloaking. The policy targets conditional serving where Googlebot specifically receives different content than human users would receive under the same request fingerprint. The test is parity within the same region, not parity across regions.
- 03.
What variants of cloaking trigger separate manual actions?
Cloaking covers standard HTML DOM mismatch. Sneaky redirects covers conditional redirects routing Googlebot and users to different URLs. Sneaky-mobile-redirects covers mobile-specific conditional redirects. Cloaked images covers Google Images-specific mismatches. AMP content mismatch covers AMP-canonical parity violations. Structured data manipulation covers markup of invisible text or irrelevant entities. Each routes to its own reconsideration template.
- 04.
How long does a cloaking reconsideration take?
Reconsideration timelines vary by category. Cloaking-class requests historically take longer than link-related requests because Google's review team verifies DOM parity manually across the affected URL set. The per-URL evidence package shortens the review; partial remediation extends it or causes rejection.
- 05.
Will my site recover full rankings after the manual action lifts?
Lifting the manual action restores indexation. The ranking recovery is a separate question that depends on the recrawl cycle, the post-remediation site structure, and whether the underlying content also requires quality work. Google does not guarantee ranking restoration; lifting the action removes the block on ranking, while the requirement to earn the rank remains.
If your Search Console flagged cloaking, sneaky redirects, AMP content mismatch, or structured data manipulation, book the diagnostic.
We run the DOM-parity audit across the affected URLs, identify every conditional serving branch, strip the logic at the application, CDN, and redirect layers, and author the reconsideration request to the matched category template with per-URL evidence.