Negative comments on Facebook ads: five kinds, and which to hide or answer

Negative comments under a Facebook ad are five different things, and only two should be hidden fast: the false accusation and the competitor's dig, because a paid audience reads them first. As of September 17, 2026, hiding from Ads Manager keeps the comment for its author and their friends and removes it for everyone else; deleting needs the ad's post. The rest is a call by kind, with a real example of each below.

Below: the five kinds with examples, the call for each, hiding from Ads Manager and who still sees it, deleting from the post, when negativity is abuse, what Moderation Assist can and cannot do, what my own negativity rule removed and should not have, the rule set to Hide, and Facebook beside Instagram.

The five kinds, from my own ads

The examples are from my own Facebook ads between August 1 and September 17, 2026, as written, names left out.

The joke and the opinion ("AI can't be trusted", September 1) sit with the price objection: a reader's view, not a claim.

The call for each

Hide the accusation and the dig, today, because an ad's readers are strangers who take a one-word "scam" at face value and never see your reply. Answer the complaint and the support question in public, because the next stranger reads the answer as the product's character; the eight-months comment above was a customer who needed a person, not a moderator. Leave the price objection and the joke; a thread with nothing but praise reads as cleaned, and a reader who thinks "$18 for coffee" is not your buyer either way. When the accusation is about a real order, it is a complaint, and the call flips to answering.

Hide from Ads Manager, and who still sees it

Meta's ad-comments page, read September 17, 2026, reaches a Facebook ad's comments through the ad's More options and Manage Facebook comments; the pop-up carries Hide comment and Report comment, with like and reply beside them. Hiding follows Facebook's Page rule, from its hide page read the same day: the comment "will still be visible to the person who wrote it and their friends" and is "hidden for everyone else", replies included. For a paid audience that is the right shape: the writer is not provoked into a second comment, and the thousand strangers never see the first.

EnlargeOpen this screen at full size

Delete from the post

Deleting is not in the pop-up. To delete, the ad needs a post: open it as the Page, click Options next to the comment, Delete, and it is gone for everyone, including the writer, with no notice and no undo. Comments on Facebook ads maps the places an ad's comments live and says which one deletes; how to delete comments on a Facebook Page has the access rule.

When negativity is abuse

A negative comment is criticism; abuse is different, and Meta's rules draw the line, not you. Its bullying policy, read September 17, 2026, says "we want to allow discussion, which often includes critical commentary", so a rude opinion about your ad stays unless you hide it, while a slur or a threat is Meta's to remove through Report. Hate and rude comments on Facebook has where the line falls and the report path.

What Moderation Assist can and cannot do here

Moderation Assist hides by criteria about the account (no friends or followers, a new profile, no picture, a repeat offender) and by keywords and links. It cannot tell "Arnaque" from "I ordered and it never came", and it hides only comments made after a criterion is added, in its own words. It is the right tool for the account with no history that posts "scam" under every ad; it is the wrong tool for a paying customer's complaint, which needs a reply and not a filter. What Moderation Assist hides and leaves has each criterion.

My own numbers, including the ones the rule got wrong

Between August 1 and September 17, 2026, 66 comments under my Facebook ads were removed as negative: 49 by my negativity rule, 15 by another tool, 2 by hand. Reading them back, three should have stayed: the eight-months complaint, the three-days setup question, and a question in Arabic about a suspended account (September 4). The rule was set to Delete, so those three were removed instead of answered. That is the honest cost of a rule with the wrong action: the accusations went, and so did three people who wanted help.

The rule, set to Hide

CommentWatchDog is an AI comment moderation tool that automatically hides and deletes spam and damaging comments on Facebook and Instagram ads and posts. The Negativity rule judges each comment by a sentence the owner writes, and the sentence on the screen below says what the count above taught: a scam accusation goes, "a real complaint about an order is not negativity, it needs a reply". Set to Hide, as on the screen below, a misjudged comment sits in the hidden view with an Unhide button for the person who reads that view; set to Delete, as mine was, it is gone; guide the negativity rule is the guide.

EnlargeOpen this screen at full size

9:41

EnlargeOpen this screen at full size

Facebook ads and Instagram ads, for negative comments

The Facebook column is drawn from the pages quoted above; the Instagram column summarises the Instagram article.

Sources

• Manage comments on your ads, Meta Business Help Centre, read September 17, 2026
• Hide a comment from a post on your Facebook Page, Facebook Help Centre, read September 17, 2026
• Delete a comment from a post on your Facebook Page, Facebook Help Centre, read September 9, 2026
• Bullying and Harassment, Meta Community Standards, read September 17, 2026
• About Moderation Assist, Meta Business Help Centre, read September 5, 2026, tried on my own profile September 8, 2026
• Best practices for comment moderation on your ads, Meta Business Help Centre, read September 10, 2026
• My own moderation log, three Facebook Pages, comments under ads published August 1 to September 17, 2026, read September 17, 2026