Answers
Where can I get historical Douyin hot search data?
Short answer
Douyin itself only shows the board as it is right now, so history has to be recorded by someone polling it. With okdata you can build your own archive: one call returns the whole board for $0.002, so hourly snapshots cost $1.44 a month and $17.52 a year.
A researcher studying how public attention moves, or a brand checking whether last spring's campaign ever touched the hot list, needs what the board looked like weeks or months ago. Douyin does not offer that. The app shows the current board; once a topic drops off, it is gone from the list. Any history that exists was recorded by someone who polled the board at the time.
Start recording now
Douyin hot search board returns the full current board, about 50 rows with rank, keyword and heat, in a single call. Store every snapshot with meta.fetched_at and you have a time series from today on.
| Frequency | Calls per month | Per month | Per year |
|---|---|---|---|
| Daily at 9:00 | 30 | $0.06 | $0.73 |
| Hourly | 720 | $1.44 | $17.52 |
| Every 10 minutes | 4,320 | $8.64 | — |
Hourly is the usual choice: topics often enter and leave within a few hours, and daily snapshots miss them.
A minimal archiver
# crontab: 0 * * * * python3 archive_douyin_hot.py
import json, os, requests
r = requests.get(
"https://api.okdata.online/v1/douyin/trends/list",
headers={"Authorization": f"Bearer {os.environ['OKDATA_API_KEY']}"},
timeout=60,
)
body = r.json()
if body["code"] == "ok":
with open("douyin_hot.jsonl", "a", encoding="utf-8") as f:
f.write(json.dumps({"at": body["meta"]["fetched_at"], "items": body["data"]["items"]}, ensure_ascii=False) + "\n")
A failed call is not charged; log it and the next hour fills the gap. From the JSONL you can compute when a word first appeared, its peak rank and how long it stayed.
Limits
- No backfill. The API cannot return the board as it was before you started recording.
- Main board only, not Douyin's category boards.
heatis Douyin's own index, useful for ranking within a snapshot and for trends over time, not an exact count of searches.
Where else history comes from
- Hot-list archive projects. Several open-source projects on GitHub save daily snapshots of Chinese hot lists, Douyin included. Check that they cover the dates and frequency you need before relying on them.
- 巨量算数 (Ocean Engine's index). Gives search interest for a keyword over time, which answers "how hot was this topic" even if it never made the board.
- Douyin's creator tools. 抖音热点宝 shows recent hot topics with analysis for creators.
If your question is about a period that has already passed, start with those. For anything going forward, your own hourly archive is the most complete. See trend research for what teams do with one.
Updated