SafeSearch and Content Policies
Last updated
SafeSearch and content policies affect search visibility differently, and the two are often conflated. SafeSearch is a preference-based filter applied at query time. Content policies govern what Google will index and show regardless of user settings. Each has distinct implications for publishers.
What SafeSearch does
SafeSearch is a filter applied by Google when serving results. When a user has SafeSearch enabled, results containing explicit sexual content or graphic violence are removed from the results page for that query.
The filtering happens at display time, not at indexing time. The underlying pages remain indexed. They are not penalised, not demoted, and not treated differently in ranking for users with SafeSearch off. A page filtered under SafeSearch is fully present in Google’s index.
SafeSearch tiers
Google operates three settings:1
- Filter: explicit images, text and links are removed from results.
- Blur: explicit images are blurred, while explicit text and links may still appear. This is the setting Google applies when Filter does not, and since 2023 it is the default state for most users who have not turned SafeSearch off.
- Off: no filtering applied. Available to signed-in adult users in supported regions.
There is no longer a “Moderate” tier; Blur replaced it. In some contexts, SafeSearch is locked at the network or account level: school networks, Google Workspace for Education accounts, and accounts belonging to users under 18. A locked SafeSearch cannot be changed by the user.
How Google classifies content
Google’s SafeSearch classifier evaluates pages based on:2
- Explicit sexual content. The primary category: nudity, graphic sexual depictions, adult content, and sex-oriented services.
- Graphic violence. Depictions of severe violence, gore, or graphic injury.
- Shocking content. Disturbing content that does not fall primarily under the above categories.
- Links, pop-ups and ads. A page that is not itself explicit can be classified for the explicit content it links to, or for explicit pop-ups and ads it displays.
Google aims to allow explicit content that has clear educational, documentary, scientific or artistic (EDSA) value, so legitimate material in those categories is less likely to be filtered even when it is graphic.2
Classification is usually per page, so a site that mixes standard and explicit content has individual pages assessed separately. But Google can also decide that an entire site “seems explicit in nature” and filter the whole site when SafeSearch is on. To prevent that, Google recommends grouping explicit content on a separate domain or subdomain, kept apart from the rest of the site.2
Marking pages as explicit
Publishers can label their own explicit pages so Google filters them reliably under SafeSearch. The main control is a meta tag in the page <head>:
<meta name="rating" content="adult">Google treats any page carrying the adult rating (or the equivalent RTA label) as explicit and filters it when SafeSearch is on, regardless of what the page actually contains. The same mechanism is the most common cause of accidental filtering: an adult rating tag, or a <video:family_friendly>no value in a video sitemap, applied to non-explicit pages will pull them out of filtered results. If clean pages are being filtered, check for a stray rating tag before anything else.2
Restricted indexation
Restricted indexation is distinct from SafeSearch filtering. It is a policy-level action that limits how or whether a site appears in results.
Adult content restriction: sites Google classifies as primarily adult content may be restricted to users with SafeSearch disabled. This is a site-level algorithmic classification, not a manual penalty.
Country-specific restrictions: content that violates local law may be removed from results in a specific country. This is not a sitewide action.
Manual actions: issued via Google Search Console when a site violates spam or quality policies. Separate from SafeSearch classification. A manual action suppresses rankings globally until resolved and reviewed.
SafeSearch filtering and manual actions are handled differently. Filtered pages are indexed and rank normally for unfiltered users. Manual action targets are actively suppressed.
Google’s content policies
Content policies cover categories Google will not index, or will remove regardless of user settings:
- Child sexual abuse material (CSAM). Automatic removal. Google scans actively and reports findings to NCMEC. No exceptions.
- Non-consensual explicit imagery. Content depicting real people in explicit scenarios without consent, removable on request.
- Certain personal information. Sensitive personal data categories can be removed on request under Google’s personal information policies.
- Legally required removals. Court orders, DMCA takedowns, and jurisdiction-specific legal requirements.
- Spam policy violations. Cloaking, hidden text, link schemes, auto-generated content with no user value.
Most sites will never encounter restrictions under any of these categories. They apply to a narrow set of harmful content and to sites that knowingly violate spam policies.
How to check if your site is affected
Page Indexing report in Google Search Console. Filters pages by status. Policy-related exclusions appear here with specific labels alongside common benign statuses such as pages crawled but not yet indexed, or pages excluded by a noindex tag.
SafeSearch test. Search for your brand or representative pages with SafeSearch enabled and disabled. Pages that appear in one and not the other are classified as explicit content.
Site: operator check. site:yourdomain.com in Google returns indexed pages. A large mismatch between known page count and indexed pages warrants investigation.
Manual Actions report in GSC. Under Security and Manual Actions. If Google has issued a manual action, it is documented here with specific guidance on what triggered it.
Reconsideration requests
Manual actions can be appealed after the underlying issue is fixed:
- Identify the specific violation from the Manual Actions report in GSC.
- Fix the issue: remove the violating content, clean up spam signals, or address the policy breach.
- Submit a reconsideration request via the Manual Actions report. Describe what was wrong and what changed.
- Google reviews the request manually. Response times range from days to weeks.
SafeSearch classification is algorithmic, but it is not final. If you believe a page is wrongly filtered, first review what may be triggering the classifier: explicit images, an adult rating tag, a <video:family_friendly>no value, surrounding copy, or explicit content the page links to. Once you have corrected the cause, Google says its classifiers can take up to two or three months to re-process the site; if the filtering persists after that, you can request a review through Google’s SafeSearch contact form. Google only overturns automatic decisions where a site has clearly been categorised incorrectly.3
Frequently asked questions
Will SafeSearch filtering hurt my rankings?
No. SafeSearch filtering is not a ranking signal. It does not affect how a page ranks for users with SafeSearch off. It is a display filter applied per user, not a quality judgement.
Can I control whether my pages are filtered?
To an extent. You can opt a page in by adding an adult rating tag, which marks it explicit. You cannot force a genuinely explicit page to bypass the filter, but if a clean page is being filtered you can remove any stray rating tag, fix the triggering content, and request a review once Google’s classifiers have re-processed the site.
Is legal adult content allowed in Google’s index?
Yes. Adult content is indexable and may appear in results for users with SafeSearch disabled. Google does not prohibit indexing legal adult content. What it restricts are illegal categories (CSAM) and policy violations (spam, cloaking), which are separate issues.
What is the difference between a manual action and SafeSearch filtering?
A manual action is a penalty issued by a Google reviewer for a specific policy violation. It suppresses your rankings globally and is documented in GSC. SafeSearch filtering is algorithmic, content-based, and affects only users with SafeSearch enabled. The two are unrelated.
