Spam, scams, and abuse in the comments of a paid Facebook ad do real damage: they undermine the social proof your creative is buying, they distract genuine prospects, and the worst of them can get your ad rejected. Hiding the bad comments quickly matters. This guide covers how to do it by hand, and how to automate it.
What hiding a comment actually does
On Facebook, hiding a comment is different from deleting it. A hidden comment stays visible to the person who wrote it (and their friends), so they rarely notice it has been hidden, but it disappears for everyone else. A deleted comment is removed entirely, which is more visible and harder to undo.
For ad comments, hiding is almost always the right tool. It quietly removes spam from public view without picking a fight with the commenter, and it does not touch the engagement count that Meta's delivery system reads. Deleting should be reserved for clear, severe cases — and even then, with care.
Hiding a comment by hand
To hide a single comment from the ad itself:
- Find the ad in your feed or in Meta Ads Manager and open its comments.
- Hover over the comment you want to hide.
- Click the three-dot menu that appears on the right of the comment.
- Choose Hide comment.
The comment is hidden immediately. You can reverse this from the same menu by choosing Unhide.
This is fine for the occasional troll. It falls apart when you are running several ads, each pulling dozens or hundreds of comments a day, often outside your working hours. By the time you see a spam comment, it may have sat under your ad for hours — in front of exactly the audience you paid to reach.
Why manual hiding does not scale on ads
Ad comments have three properties that make manual moderation painful:
- Volume. A single well-funded ad can attract hundreds of comments. Multiply that across a campaign and you are reading comments all day.
- Timing. Spam often lands within minutes of an ad going live, and the highest-spend hours are not always the hours you are awake.
- Dark posts. Many ads run as dark posts — creatives that never appear on your Page timeline — so they are easy to forget to check at all.
The result is that the comments you most want to remove are the ones that sit unmoderated the longest.
Automating it with ROAS Shield
ROAS Shield is built specifically for moderating Facebook and Instagram ad comments. It connects to your Pages through Meta's official Graph API — no scraping, no browser automation — and processes each new comment as it arrives.
Here is what it does with a new comment:
- Classifies it (spam, abuse, off-topic, neutral, or buyer-intent).
- Applies your moderation rules — for example, hide anything classified as spam.
- Protects the comments you would never want to lose: a buyer-intent signal vetoes any automatic delete and demotes it to a hide at most.
Auto-deletion is deliberately hard to switch on. It is schema-blocked unless you explicitly opt in and the model's confidence is very high, and even then it is restricted to workspace owners and admins. The safe default is to hide, not delete — see ad comment moderation for the wider picture.
Because ROAS Shield reacts the moment a comment is posted, spam is gone in seconds rather than hours, and you are not the one reading it.
What it deliberately leaves alone
A blunt keyword filter hides too much. The comment that reads "how much is shipping to Canada?" is not spam — it is a customer trying to give you money. ROAS Shield treats those buyer-intent comments as the most valuable thing in the thread and routes them to you to convert, instead of burying them. If you want to reply to those automatically too, see our guide on auto-replying to ad comments.
Pricing
ROAS Shield plans start at £19/month for the Starter tier (10,000 comments/month) and go up to £199/month for Scale (500,000 comments/month). Every paid tier includes automatic hide rules. Full details are on the pricing page, and common questions are answered in the FAQ.