TikTok scraper API: profiles, videos, comments and hashtags

TikTok is where many trends, products and creators break out first, which makes its public data valuable for influencer marketing, brand monitoring, music analytics and social research. TikTok offers official APIs, but access is limited to specific programs and use cases, and approval can take time. For teams that need public profile and video data now, a scraper API is often the pragmatic route.

This guide walks through the TikTok endpoints one by one: what each returns, when to use it, and how to combine them.

The endpoints at a glance#

All of them take a small JSON body, return the same response envelope and use the same error codes, so once one works, the rest are a copy-paste away.

Profiles: vetting creators#

The profile endpoint is the starting point for creator research. Send a username, an @handle or a profile URL:

curl -X POST https://api.scrapeeverything.com/v1/tiktok/profile \
  -H "Authorization: Bearer $SCRAPEEVERYTHING_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"username": "@creator_handle"}'

You get follower_count, following_count, like_count (total likes received), video_count, is_verified, biography, external_url, region and more. Two derived numbers are useful for quick vetting:

• Likes per video: like_count / video_count. A rough signal of how well the average video lands.
• Followers to following ratio: very low ratios can indicate follow-for-follow growth.

Neither is a verdict on its own. Combine them with the performance of recent videos, which is what the next endpoint is for.

Recent videos: measuring real reach#

The videos endpoint returns the profile summary plus the newest videos with play_count, description and cover images. Follower counts say how many people could see a video; play counts on recent videos say how many actually did. For campaign planning, the median play count of the last 12 videos is a more honest number than the follower count.

async function tiktok(path, body) {
  const res = await fetch(`https://api.scrapeeverything.com/v1/tiktok/${path}`, {
    method: "POST",
    headers: {
      Authorization: `Bearer ${process.env.SCRAPEEVERYTHING_API_KEY}`,
      "Content-Type": "application/json",
    },
    body: JSON.stringify(body),
    signal: AbortSignal.timeout(90_000),
  });
  const json = await res.json();
  if (!json.success) throw new Error(`${json.code}: ${json.message}`);
  return json.data;
}

const { posts, follower_count } = await tiktok("posts", { username: "creator_handle", cache_days: 1 });
const plays = posts.map((p) => p.play_count).sort((a, b) => a - b);
const median = plays[Math.floor(plays.length / 2)] ?? 0;
console.log({ follower_count, median_plays: median, views_per_follower: median / follower_count });

Single videos: campaign reporting#

When a creator publishes a sponsored video, you want its numbers over time: plays, likes, comments, shares and saves. The video endpoint takes the URL and returns all of them, plus the author and the sound used. Schedule it daily for the first two weeks after publishing and you have a clean performance curve for your report.

Comments: what the audience actually says#

The comments endpoint returns top-level comments with text, like_count, reply_count, created_at, is_pinned, language and the comment author. Typical uses:

• Sentiment analysis on sponsored content.
• Finding product questions and complaints.
• Checking whether engagement looks organic (repetitive generic comments are a warning sign).

import os
import requests

res = requests.post(
    "https://api.scrapeeverything.com/v1/tiktok/comments",
    headers={"Authorization": f"Bearer {os.environ['SCRAPEEVERYTHING_API_KEY']}"},
    json={"url": "https://www.tiktok.com/@creator_handle/video/VIDEO_ID", "limit": 50},
    timeout=90,
)
data = res.json()["data"]
for c in sorted(data["comments"], key=lambda c: c["like_count"], reverse=True)[:10]:
    print(c["like_count"], c["text"])

Transcripts: making videos searchable#

Many TikTok videos carry captions, either added by the creator or generated automatically. The transcript endpoint returns them with timestamps, which is a low-cost way to know what a video says without watching it. It pairs well with LLM summarization and brand mention detection. The response has the same shape as the YouTube Transcript API, so one pipeline can process both.

Hashtags and sounds: tracking trends#

The hashtag endpoint returns the total video_count and view_count for a hashtag. A single snapshot is just a number; daily snapshots turn it into a trend line. The difference between two days is the growth, and growth relative to size tells you whether a trend is accelerating or fading.

The sound endpoint does the same for audio: title, artist, duration and whether it is an original sound. Music marketers use it to follow how a track spreads on TikTok after release.

Putting it together: a creator research pipeline#

1. Start from a list of candidate usernames (from search, your CRM or creator submissions).
2. Fetch each profile with cache_days: 7. Drop private accounts and those below your follower threshold.
3. For the remaining creators, fetch recent videos and compute median plays per video.
4. For the shortlist, read comments on two or three recent videos to check audience quality.
5. After the campaign launches, track the sponsored videos daily with the video endpoint.

At 1 credit per call, vetting a creator with this pipeline costs a handful of credits. Failed requests are refunded and cached repeats are free, so re-running the pipeline every week stays cheap. See pricing for pack sizes.

Limits worth knowing#

• Only public data: private accounts return their public profile fields, not their videos.
• The videos endpoint returns the newest videos visible publicly, not the full history of an account.
• Counts like plays and likes are as TikTok shows them at the time of the request.
• Comment endpoints return top-level comments; replies are counted in reply_count.

Next steps#

Browse every field in the TikTok API reference, then sign in to get 100 free credits and try the endpoints in the playground. If you are deciding between providers, our pricing comparison lays out how the main options charge.