tiktok.sound_trends

derived · one row = one sound with > 3 videos

Where each sound is (stage) and how fast it moves: 1/7/30-day change and growth %.

Coverage: Growth % = change / previous with a 100-video floor. NULL when a day is missing.
Refresh: Daily, after sound_daily closes the day.
Sorted by stage, sound_id. Filters on the leading columns are fastest.

Example

SELECT *
FROM tiktok.sound_trends
LIMIT 100

Column	Type	Description
`sound_id`	`UInt64`
`as_of`	`Date`
`title`	`String`
`artist`	`String`
`owner_creator_id`	`UInt64`
`is_original`	`UInt8`
`has_custom_title`	`UInt8`
`matched_type`	`LowCardinality(String)`
`videos_now`	`UInt32`
`change_1d`	`Nullable(Int64)`
`change_7d`	`Nullable(Int64)`
`change_30d`	`Nullable(Int64)`
`growth_pct_1d`	`Nullable(Float32)`	change / greatest(previous, 100) * 100
`growth_pct_7d`	`Nullable(Float32)`
`growth_pct_30d`	`Nullable(Float32)`
`stage`	`LowCardinality(String)`	Size bands: emerging < 10k videos, accelerating 10k-100k, mainstream 100k-1M, saturated >= 1M.
`momentum`	`LowCardinality(String)`	rising: growth_pct_7d >= 10 · falling: < 0 · flat: otherwise · ‘’ no 7d base
`first_seen`	`DateTime`