Tag videos automatically with metric rules and AI rules.
Workflows tag videos for you. Create one in the Workflows tab of Tags, and it runs on every new video as it syncs. Manual tagging stays useful for one-off decisions; workflows handle the labels you would otherwise apply hundreds of times.
There are two rule types:
Apply tags or exclude videos when they match filters such as views, engagement rate, platform, publish date, hashtags, or caption keywords. Instant and free.
The AI watches each video and decides which of your selected tags apply. Uses credits.
A metric rule checks each video against conditions you define, such as minimum views, engagement rate, platform, publish date, hashtags, or caption keywords, and applies your tags when they match. A metric rule can also exclude matching videos from analytics, which replaces manual exclusion lists for patterns you already know, like spam hashtags or off-topic posts.
An AI rule sends each qualifying video to an AI that watches it and picks which of your selected tags apply. The tag name and description are its instructions, so write descriptions like criteria: "Apply when the creator unboxes a physical product on camera." The optional context field tells the AI what your product and videos are about.
By default the AI assigns any number of the selected tags, including none. Switch the classification mode to Restrict to set a minimum and maximum per video, for example exactly one tag when every video should land in a single bucket.
Before a rule goes live, use the free test run to check the verdict on one of your own videos. Conditions work on AI rules too: they narrow which videos the AI evaluates, which lowers the cost.
Each video the AI evaluates costs credits, and a video is only evaluated once per rule. New videos are processed as they sync. To cover videos already in your analytics, turn on the backfill when you create the rule. The preview shows how many videos match and the exact credit cost before you confirm anything.
Workflows are only as good as the tag library behind them, so start with clear names and descriptions in Tags. Once tags land on videos, filter Tracking analytics by tag. If you still organize reporting with projects, the migration from projects guide shows how to move those labels into tags first.
