Contextual moderation means judging content by its surroundings, not by the words or image alone. Who posted it, who it's aimed at, what conversation it's part of, and what community norms apply all feed into the decision.

‍

‍Why it matters

The same words can be harmless in one place and abusive in another. A slur quoted in a news report isn't the same as one aimed at a person. Friends joking in a gaming chat aren't the same as a stranger sending the identical message to a teenager. Systems and reviewers that only see a single post in isolation get these wrong in both directions, removing harmless content and missing harmful content that looks mild on its own.

‍

‍How teams apply it

In practice it means giving reviewers the thread and account history instead of a lone post, and writing guidelines with context-specific examples, not just lists of banned terms. It also means having reviewers who know the language, slang, and culture of the content they're judging. Context takes longer to assess, so teams usually save the full review for cases where automated checks and quick reviews disagree.