TikTok product research: hot list → category → detail → reviews
Four endpoints chain together: tiktok_shop_hot_products for the bestseller list per region, tiktok_shop_category_products to widen the pool, tiktok_shop_product_detail for price/stock, and tiktok_shop_product_reviews to spot review manipulation. ¥0.2 per call.
Which endpoints to call
| Endpoint | Price | Calls | Subtotal | Purpose |
|---|---|---|---|---|
tiktok_shop_categories | ¥0.2 | ×1 | ¥0.2 | Category tree (one-off) |
tiktok_shop_hot_products | ¥0.2 | ×3 | ¥0.6 | Bestseller list per region, paginate with IDs you already have |
tiktok_shop_category_products | ¥0.2 | ×4 | ¥0.8 | Widen the candidate pool by category |
tiktok_shop_product_detail | ¥0.2 | ×1 | ¥0.2 | Price, sales, variants |
tiktok_shop_product_reviews | ¥0.2 | ×2 | ¥0.4 | Real buyer feedback, to detect fake volume |
| per candidate product (1 category tree + 3 hot-list + 4 category + 1 detail + 2 reviews = 11 calls) | ¥2.2 | per full run | ||
| 200 products | ¥440 | 200 × ¥2.2 | ||
Working code skeleton
Runs as-is once you set TOKEN. Failed and empty calls are skipped, so they cost nothing.
import time, requests
TOKEN, API = "fw_xxx", "https://finewisdom.top/api"
REGION = "MY" # ID/JP/MX/MY/PH/SG/TH/VN
def call(method, **kw):
r = requests.get(API, params={"token": TOKEN, "method": method, **kw}, timeout=60).json()
return r["data"] if r.get("status") == "success" else {}
def hot_products(region=REGION, pages=3):
seen, out = [], []
for _ in range(pages):
d = call("tiktok_shop_hot_products", region=region,
exclude_product_ids=",".join(seen) if seen else "")
items = d.get("list") or d.get("products") or []
if not items:
break
out += items
seen += [str(i.get("product_id")) for i in items]
time.sleep(1.1)
return out
def enrich(pid):
d = call("tiktok_shop_product_detail", product_id=pid, region=REGION)
rev = call("tiktok_shop_product_reviews", product_id=pid, region=REGION)
return {"product_id": pid, "detail": d, "reviews": (rev.get("list") or [])[:40]}
Gotchas worth knowing
regionmust be passed — the default is not the country you want. The US isUS; Southeast Asia must be pulled per market, because bestseller lists differ completely.- To paginate the bestseller list you must pass
exclude_product_ids(comma-separated IDs already collected), otherwise you get the same products again and pay twice for nothing. - To detect review manipulation:
tiktok_shop_product_reviewsreturns timestamps and text. Products whose reviews cluster in a few days with near-identical wording should be dropped.
FAQ
Can I get GMV?
No. TikTok Shop only exposes public sales signals and reviews. Exact GMV lives in the merchant backend; no third party can obtain it.
Bestseller list or category browse?
Use the bestseller list to find what is already selling, then category browse to find cheaper alternatives in the same niche. Doing both beats either alone.