How-to5 min read

How to block words in Facebook and Instagram comments

Set up keyword blocking with Instagram Hidden Words and Facebook Page moderation settings, then fix the false positives raw blocklists create.

By CommentProtect Team · Published July 3, 2026

Keyword blocking is the fastest native win in comment moderation: decide which words, phrases, and emojis you never want under your content, and let the platform hide matching comments automatically. Facebook and Instagram both support it, with slightly different mechanics.

Blocking words on Instagram

Instagram's Hidden Words feature has two parts: a platform-maintained filter for offensive content you can toggle on, and a custom list where you add your own words, phrases, and emojis. Comments that match are hidden from public view rather than deleted, so you can review and unhide mistakes. Hidden Words lives in the app's settings; search for it by name if the menus have moved.

Blocking words on Facebook Pages

Facebook Pages have keyword moderation in the Page settings: comments containing words you block are hidden automatically. Pages also offer a profanity control backed by Meta's own lists. Like Instagram, matched comments are hidden — still visible to their authors — not removed.

Where raw keyword lists break down

  • Creative spelling and leetspeak slide past exact-match lists.
  • Multilingual audiences need every variant in every language, forever.
  • Innocent uses of a blocked word get hidden too — false positives cost real conversations.
  • Lists rot: yesterday's scam phrasing is not tomorrow's.

Pair keywords with AI rules

Keyword lists are precise but brittle; AI moderation is the complement. AI profanity and negativity rules judge meaning rather than exact strings, which catches misspellings and other languages without you enumerating them. A whitelist protects the terms that are innocent in your context, and hidden comments stay reviewable so a human can correct any call.

A setup that holds up

  • Turn on the native offensive-content filters on both platforms.
  • Add a short custom list: scam phrases, contact-detail bait, and terms specific to your brand's risks.
  • Layer AI profanity and negativity rules for everything the list cannot spell.
  • Auto-hide URLs, emails, and phone numbers — most spam needs a link to work.
  • Review the hidden queue weekly and prune rules that misfire.
FAQ

Common questions

Are comments with blocked words deleted?

No. On both Facebook and Instagram, keyword-filtered comments are hidden, not deleted. The author still sees their own comment, and you can unhide anything caught by mistake.

How many words should I block?

Start small: obvious scam phrases, contact-detail bait, and the handful of terms specific to your brand's risks. Long lists multiply false positives. Let AI rules handle profanity and negativity so the keyword list stays short and precise.

Do blocked-word filters work in other languages?

Only if you add each language's variants yourself, which is why multilingual audiences outgrow keyword lists quickly. AI moderation rules that evaluate meaning across supported languages are the more durable approach.

Put this into practice

Request controlled access when you are ready to evaluate CommentProtect with your Meta workflow.