EnsembleData alternative: ScrapingBot vs EnsembleData

EnsembleData is a social media data API built around unit quotas that reset every day. If your jobs run evenly through the month, that model works well. If they arrive in bursts, or if you need search and e-commerce data next to social data, the comparison looks different.

The short version

ScrapingBot vs EnsembleData at a glance

How the per-1,000 numbers work: Wood gives 1,500 units × 30 days = 45,000 units for $100, so a unit costs $0.00222. That holds only if you spend the whole allowance every day. ScrapingBot: $49.99 ÷ 275,000 credits = $0.00018 per credit.

Daily units vs monthly credits

The pricing model matters more than the headline price. Picture a team that refreshes 2,000 creator profiles every Monday and backfills 30,000 sound videos when a client signs on.

• On a daily quota, Monday's 2,000 calls exceed Wood's 1,500 units, so part of the refresh waits until Tuesday, and the units from the quiet days in between are lost. The 30,000-video backfill (about 3,000 sound pages) takes two days on Wood, or one day on Bronze at $200.
• On a monthly pool, both jobs run when you start them, limited only by concurrency (10 parallel requests on Starter). The month's total, 2,000 × 4 + 3,000 = 11,000 calls, is 11,000 credits, well inside Starter's 275,000.

The reverse case: a pipeline that runs 40,000 one-unit calls every single day. That is 1.2 million calls a month, which fits Platinum ($1,400) at about $0.93 per 1,000, or ScrapingBot Pro ($249.99, 3.3M credits) at about $0.08 per 1,000 TikTok calls. Run your own numbers with the unit table in EnsembleData's docs.

Coverage

Both cover the core TikTok jobs: profiles, a user's posts, sounds and the videos that use them, comments, replies, followers, following and keyword search. EnsembleData has more TikTok endpoints (20 against 11), including hashtag posts and full-search variants, and its followers endpoints take id plus secUid. ScrapingBot's followers and following endpoints return one page of up to 50 accounts.

EnsembleData also covers YouTube, Threads, Reddit, Twitch, Twitter and Snapchat. ScrapingBot doesn't. In exchange, ScrapingBot's credits also buy Google results (web, news, shopping, places, maps and reviews), Amazon product data and rendered web pages.

Migrating from EnsembleData

EnsembleData takes a GET with the token in the query string. ScrapingBot takes a POST with the key in a header and the endpoint in the body. The videos that use a sound, before and after:

# Before: EnsembleData
import requests

res = requests.get(
    "https://ensembledata.com/apis/tt/music/posts",
    params={"music_id": "7682907287056337667", "cursor": 0, "token": ENSEMBLE_TOKEN},
    timeout=60,
)
data = res.json()["data"]
items, next_cursor = data["aweme_list"], data["nextCursor"]

# After: ScrapingBot
import requests

res = requests.post(
    "https://scrapingbot.io/api/v1/tiktok",
    headers={"x-api-key": SCRAPINGBOT_KEY},
    json={"endpoint": "/music/posts", "params": {"music_id": "7682907287056337667"}},
    timeout=60,
)
data = res.json()["data"]
items, next_cursor, more = data["videos"], data["cursor"], data["hasMore"]
# keep paging with params["cursor"] = next_cursor until more is 0

Each ScrapingBot video is a flat object with video_id, title, play_count, digg_count, comment_count, create_time and author, so map your fields once. A profile from /user/info returns data.user and data.stats, the same top-level split as EnsembleData's tt/user/info.

One difference to plan for: EnsembleData's comment endpoints take an aweme_id, while ScrapingBot's /comment/list takes the video URL. The comments export guide shows the full loop, including replies.

Common questions