How to Filter Spam Comments on Facebook
Facebook Pages attract spam like a magnet: "DM me to grow your account", fake giveaways, crypto bait, and competitor links. This guide walks through the native tools Meta gives you, where they fall short, and how to close the gaps.
Step 1: Turn on Meta's built-in keyword filter
Open your Page, then go to Settings → Privacy → Public post comments. Meta lets you add up to 1,000 keywords that will automatically hide matching comments. Start with the obvious ones:
- Money-bait terms: "bitcoin", "crypto", "forex", "investment", "loan"
- Growth-spam phrases: "dm me", "check my page", "grow your followers"
- Profanity and slurs specific to your audience's languages
- Your competitors' brand names, if they're being dropped in comments
This filter is literal, not contextual. It will miss "in.vestm.ent" or emoji substitutions, and it will hide legitimate comments that happen to contain "loan" — a real problem for finance pages.
Step 2: Configure links, photos, and repeated comments
Under the same Privacy menu, enable block links and block photos for comments from people who don't follow you. Most comment spam includes an external link or an image, so this single toggle removes a large chunk of it. Trade-off: genuine customer questions that include a screenshot get held too.
Step 3: Use the profanity filter with care
Meta's profanity filter has three levels: Off, Medium, Strong. Medium catches most abusive language while permitting mild frustration; Strong will hide comments that say "this pricing is shit" — which is negative, but often feedback you actually want to read.
Step 4: Set up moderation assistants and blocked words per Page
If multiple people manage your Page, add moderators rather than sharing login credentials. Moderators can hide, delete, and ban without accessing billing or ad settings. Audit moderator access quarterly.
Where Facebook's native tools break down
- No sentiment analysis. A sarcastic "great, another fee" reads as praise to a keyword filter.
- No cross-platform view. Instagram and YouTube need separate manual work.
- No ad comment isolation. Ad comments live in Ads Manager, not Page settings.
- No automatic replies. Every "how much?" question waits for a human.
- No multilingual context. Filters are literal strings; they don't understand Bahasa Malay spam patterns unless you type them out.
Step 5: Add AI filtering on top of native rules
Social Kirin reads the full comment, the post context, and prior behaviour before deciding. Spam that survives keyword filters — obfuscated links, emoji-only crypto bait, "collab?" scams — gets caught, while legitimate comments containing trigger words get through. Auto-hide actions run in seconds, so your audience never sees the spam in the first place.
For a full feature breakdown, see Facebook comment moderation.
Frequently asked questions
Does hiding a comment notify the person who wrote it?
No. The commenter still sees their own comment and can't tell it was hidden, though others won't see it. This is why hiding is usually better than deleting for borderline cases — deleting invites a confrontation.
How many keywords can I block?
Facebook allows up to 1,000 blocked keywords per Page. Instagram allows far fewer, which is one reason cross-platform tooling exists.
Can I filter comments on boosted posts and ads?
Ad comments are managed separately in Ads Manager, and Page-level keyword rules do not always apply. If you run paid campaigns, read protecting ad comments from trolls.