Solutions · Data and research teams
A TikTok data API that answers the question, not the scrape.
Most teams that search for a TikTok API want a series: how much a subject is discussed, by whom it is watched, and which way it leans. We return that series, with its sample size.
Series, not posts
Query by region, topic, entity and date range. Each point carries its share, its stance measures, n and a confidence interval.
Nothing to maintain
No crawler, proxy pool or parser on your side, and no breakage when the platform changes a page.
The same records three ways
REST for applications, Parquet and CSV drops for research, and the terminal for people who do not write code.
Request
curl -sS https://api.clamordata.com/v1/series \
-H "Authorization: Bearer $CLAMOR_API_KEY" \
-G -d dataset=social-discourse-index \
-d region=US -d topic=housing \
-d from=2026-08-01 -d to=2026-09-01Endpoints, response shape and delivery formats are on the API & delivery page.
| Scraper or unofficial API | Clamor Data | |
|---|---|---|
| What you get | Raw posts, profiles and media | Measured aggregates on a fixed schema |
| Sampling | Whatever a hashtag or search returns | A random sample of uploads |
| Meaning | You build the classifiers | Topic, entity and stance already measured, versioned |
| Personal data | You hold it and answer for it | None delivered: no profiles, no post-level rows |
| Upkeep | Breaks with every platform change | Ours |
Is this TikTok's official API?
No. TikTok's Research API is an official programme for approved academic and non-profit researchers and is limited to non-commercial use. Clamor Data is an independent data provider: we measure public videos and deliver aggregates to commercial and research customers.
Can I download videos, profiles or comments through it?
No. The API returns aggregates with a minimum cell size. We do not deliver videos, creator profiles, individual comments or transcripts.
What if I need a topic or entity you do not track?
Custom slices are built against a specific research question and delivered on the same schema. Ask for it in the sample request.
Request a sample dataset.
Tell us the market and the question. We return a real slice, with schema, sample sizes and confidence intervals included. Not a brochure.