Douyin Video Analytics Scraper
Pricing
from $8.00 / 1,000 results
Douyin Video Analytics Scraper
Get performance analytics for one public Douyin video by URL or aweme_id: the frequent-word breakdown of its comment section, and the demographic profile -- region, gender, age, and device -- of viewers who liked it. This data normally needs a logged-in Douyin creator account to see.
Pricing
from $8.00 / 1,000 results
Rating
0.0
(0)
Developer
ToolzerHub
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
8 hours ago
Last modified
Categories
Share
Get performance analytics for one public Douyin video: the frequent-word breakdown of its comment section, and the demographic profile — region, gender, age, and device — of viewers who liked it. This data normally needs a logged-in Douyin creator account to see. Pick a mode with scraperType.
Modes
scraperType | Returns |
|---|---|
videoCommentWords (default) | Frequent-word breakdown of the video's comments |
videoAudience | Demographic profile of viewers who liked the video |
Input
| Field | Use it for |
|---|---|
aweme_id | A Douyin video URL, or the numeric video ID it contains. Required for both modes. |
option | videoAudience only. Integer 1–9, default 4. |
addonVideoDetails | Fetches full video detail alongside the analytics — one extra request and one extra charge. |
Only works for videos currently on a Douyin board
Douyin only serves both videoCommentWords and videoAudience for videos that are currently sitting on one of its hot.douyin.com boards, inside that board's viewable window. A video that has aged off — even one that was popular last month — returns nothing. Pull aweme_id values from a currently-live board rather than an arbitrary catalog of past videos if you want these two modes to return data.
videoCommentWords is one row, not one row per word
The whole comment word cloud — roughly 80 words on a well-commented video — comes back from a single upstream request. This Actor keeps that as one row holding a comment_words array, rather than exploding it into one row per word, for the same reason as its account-level counterpart: a row means one request's worth of work, consistently.
Output
| Field | Mode | Contents |
|---|---|---|
comment_words, aweme_id, source_aweme_id, video_detail | videoCommentWords | The whole word cloud, one row |
portrait, portrait_tgi, option, aweme_id, source_aweme_id, video_detail | videoAudience | Demographic breakdown of viewers who liked the video |
Questions
I ran this against a video I know is popular and got nothing back — why? Almost certainly because the video isn't currently on a hot.douyin.com board. See the note above — both endpoints scope to videos inside a board's active window, not any public video by ID.
Why is videoCommentWords one row instead of ~80?
By design — see the note above. Read the individual words out of comment_words.
Where do I find a valid aweme_id to test with?
The video-performance modes of the Trending Scraper — videoBoard, lowFollowerHits, highLikeRate, and the like — return aweme_id for videos that are, by definition, currently on a board.
Related Actors
| Actor | Purpose |
|---|---|
| Douyin Trending Scraper | Find an aweme_id currently on a board before running this Actor against it |
| Douyin Video Scraper | Caption, sound, hashtags, and engagement stats for the same video |
| Douyin Account Analytics Scraper | The same kind of analytics, scoped to a whole account instead of one video |
