Your ads only appear on safe content
Every page on Urban Dictionary is scored, classified, and gated before any ad can run on it.
Content moderation
Urban Dictionary is a user-generated content site. We apply two layers of moderation — one before content is published, and another before ads are served.
Layer 1
Pre-publication review
Every user submission is evaluated before it goes live. Content is rejected if it contains:
- Attacks on private individuals
- Group-based hate or threats
- Content that endangers safety
- Spam, nonsense, or low-quality submissions
Content that does not meet editorial standards is never published.
Layer 2
Advertiser safety scoring
Every published definition is analyzed by an AI content moderation system and scored across 13 harm categories. The resulting score determines whether ads can appear on that page.
Pages that have not yet been scored default to the most restrictive tier — no ads are served on unreviewed content.
Three-tier classification
Each page is classified into one of three tiers based on the highest-severity content on the page. The worst definition on the page determines the tier for the entire page.
Safe
Full ad inventory. Content meets all advertiser safety standards across all 13 categories.
Examples: "GOAT," "lowkey," "stan"
Sensitive
Limited ad selection. Content has elevated scores in one or more categories but does not reach the highest severity.
Cautious advertisers can exclude this tier
Mature
Content exceeds safety thresholds. Also the default for any page that has not yet been scored. Advertisers can exclude this tier entirely.
Excluded by most brand advertisers
13 harm categories scored
Every definition receives a probability score (0.0–1.0) for each category. These scores feed a trained classifier that determines the page tier.
Sexual Content
2 subcategories
Harassment & Hate
4 subcategories
Violence
2 subcategories
Self-Harm
3 subcategories
Illicit Activities
2 subcategories
Ad placement controls
The safety tier is passed as a signal to the ad server on every page load. Advertisers and their DSPs can use this to control where their ads appear.
Page-level signals
The safety classification is available as a key-value in Google Ad Manager on every ad request. Buyers can target or exclude by tier.
Unreviewed content defaults to "mature"
Any page without a completed safety review is automatically classified as "mature" — the most restrictive tier. Advertisers who exclude mature content will never appear on unreviewed pages.
Per-page granularity
Classification is per page, not per site. A word like "brunch" is classified independently from a word like "wasted." Your ads only appear next to content that meets the tier you select.
Continuously updated
New submissions are scored as they're published. The classification system runs continuously — it's not a one-time review.
Questions about our brand safety?
We're happy to walk through our moderation system, share threshold details, or discuss custom requirements.