Content Moderation
for Creator Video.
A creator promoting you can get struck for a health claim they said out loud without thinking. Hubfluence transcribes every video they posted in the last 30 days and flags your keywords, so the video gets edited before TikTok acts on it.
@ava.lumen"cures acne"Transcript0:14MessageDismiss
@noor.beautyMissing "#ad"CaptionpostMessageDismiss
@bree.skintok"FDA approved"Transcript0:31MessageDismissFlags come from the spoken words, not just the caption. Message the creator or dismiss it.
Catch the violation before
TikTok does.
TikTok restricts medical and health claims, guaranteed results, and plenty of category-specific language. A creator who says one on camera can lose reach, lose the video, or take a strike on their account for promoting you.
Scans the spoken words
Every TikTok Shop video your creators posted in the last 30 days gets transcribed and scanned, so a claim made on camera is caught even when the caption is clean.
Your keywords and phrases
You define what gets flagged. Add the medical claims, the absolutes, and the phrases your category cannot use, or flag a post when a required term like your #ad disclosure is missing.
A queue, not an alert storm
Flags land in one review list with the matched term and the field it came from. Dismiss what is fine and act on what is not.
Fix it with a DM
Message the flagged creators in one action, one DM each, sent from your own TikTok account asking for a caption edit or a re-record before the video gets actioned.
Write the rule,
read the transcript, act.
Content Moderation runs on the creator videos already flowing into Hubfluence. Three steps, no new tool. Scroll to walk it.
Say what you cannot say.
Write the terms you never want spoken, and the terms that must be there. A banned-claim rule flags a word when it appears. A disclosure rule flags a post when the required term is missing. Pick whether each rule reads the transcript, the caption, the hashtags, or all three.
Say what you cannot say.
Write the terms you never want spoken, and the terms that must be there. A banned-claim rule flags a word when it appears. A disclosure rule flags a post when the required term is missing. Pick whether each rule reads the transcript, the caption, the hashtags, or all three.
Pick which fields a rule reads. A rule that skips the transcript never pays for one.
Read what was actually said.
Hubfluence transcribes every TikTok Shop video your creators posted in the last 30 days and scans the spoken words against your rules. A claim a creator says out loud but never writes in the caption is exactly the one that gets a brand in trouble, and exactly the one a caption-only check misses.
@ava.lumenNight routine, 3 week update1 flagA caption-only check never sees this. The claim was spoken, not written.
Message the creator, or dismiss it.
Open flags land in a review queue with the matched term and where it came from. Dismiss the ones that are fine. For the rest, send one DM per creator from your own TikTok account asking for the fix, and the flag moves out of the queue.
One DM per creator, sent over your own TikTok account.
@ava.lumenDM sent@noor.beautyDM sent@bree.skintokDM sentA flag you message moves out of the queue. A flag you dismiss closes without a message.
Start with the claims
that scare your lawyer.
Most brands get real value from two rules: the claims that put a creator's account at risk, and the disclosure every paid post needs.
List your banned keywords.
Add the exact words and phrases your category cannot use, the medical and health claims, the cure and guarantee language, into one rule that scans transcripts and captions.
Add the disclosure rule.
Set your required term, usually the #ad tag, as a flag-if-missing rule over captions and hashtags so an undisclosed paid post surfaces on its own.
Work the queue.
Review the flags, dismiss the false positives, and DM the creators who need to edit a description or re-record before TikTok notices.
Content Moderation
questions, answered.
Content Moderation scans the TikTok Shop videos your creators posted in the last 30 days against keywords and phrases you define, and flags the ones that match. It reads the spoken transcript, the caption, and the hashtags, so a risky claim made on camera is caught even when the caption says nothing. The point is to catch a problem while you can still fix it, rather than after TikTok has actioned the video.
TikTok restricts medical and health claims, guaranteed outcomes, and other category-specific language. When a creator breaks one of those rules promoting your product, TikTok can suppress the video, remove it, or penalize the creator's account, and the creator carries that penalty for your campaign. Flagging the video early means the creator can edit the description or re-record before TikTok acts.
Yes, and you should. Every category has its own danger list, so you write the rules. Add the exact terms you cannot risk, whether that is cure, heal, FDA approved, guaranteed results, or anything specific to your product, and choose whether each rule scans the transcript, the caption, the hashtags, or all three.
It transcribes the audio. That is the point of the feature. The claims that get a creator penalized, curing something, guaranteeing a result, calling a product FDA approved, are usually spoken in the video and never typed in the caption, so a caption-only check misses them entirely.
Yes. A rule can flag a required term when it is absent, not only a banned term when it is present. Set your disclosure tag as a required term over captions and hashtags and any paid post that went out without it shows up in the review queue.
Any combination of the transcript, the caption, and the hashtags. You choose per rule. A rule that only needs the caption skips transcription entirely, which keeps scanning cheap when you do not need the audio.
The flag lands in a review queue with the matched term and the field it came from, so you can see why it fired. Dismiss it if it is fine. If it is not, select the flagged creators and send a message, one DM per creator, from your own TikTok account.
The last 30 days of your creators' TikTok Shop videos. That window covers the content still actively selling, which is the content worth fixing.
Content Moderation is included on the Business plan and above.
Sure none of your creators
said something risky?
Book a walkthrough and we will scan a window of your creators' videos, show you what the transcripts actually say, and what your keywords would have flagged. No deck.
Most flags come from the audio, which a caption check never reads.