获胜的 TikTok 产品遵循可预测的模式：话题标签激增，创作者采用率随之上升，销量在 72 小时内激增。本教程使用 Scavio TikTok API 以编程方式检测该模式 - 扫描主题标签以了解参与速度、与视频内容交叉引用，并在趋势达到顶峰之前根据病毒潜力对产品进行评分。
前置条件
- Python 3.11+ 或 Node.js 20+
- 来自 https://scavio.dev 的 Scavio API 密钥
- 对 TikTok 内容趋势的基本了解
- 可选：用于跟踪结果的电子表格或数据库
操作指南
步骤 1: 搜索产品信号的热门标签
使用 Scavio TikTok 主题标签端点查找最近在产品相关领域增长迅速的主题标签。过滤包含产品意图关键字的主题标签。
Python
import httpx
from datetime import datetime
SCAVIO_API_KEY = "your-api-key"
BASE_URL = "https://api.scavio.dev/api/v1/tiktok"
PRODUCT_NICHES = [
"kitchen gadget",
"beauty tool",
"home organization",
"fitness accessory",
"phone accessory",
"cleaning hack",
]
async def search_trending_hashtags(niche: str) -> list[dict]:
async with httpx.AsyncClient(timeout=15) as client:
resp = await client.post(
f"{BASE_URL}/hashtag/search",
headers={"Authorization": f"Bearer {SCAVIO_API_KEY}"},
json={"query": niche, "limit": 20}
)
resp.raise_for_status()
hashtags = resp.json().get("hashtags", [])
# Filter for hashtags showing growth
trending = [
h for h in hashtags
if h.get("view_count", 0) > 100000
]
return sorted(trending, key=lambda h: h.get("view_count", 0), reverse=True)步骤 2: 获取每个热门主题标签的热门视频
对于每个有前途的主题标签，提取表现最好的视频来分析内容中出现的产品以及创作者如何呈现它们。
Python
async def fetch_hashtag_videos(hashtag_name: str, limit: int = 10) -> list[dict]:
async with httpx.AsyncClient(timeout=15) as client:
resp = await client.post(
f"{BASE_URL}/video/search",
headers={"Authorization": f"Bearer {SCAVIO_API_KEY}"},
json={"query": hashtag_name, "limit": limit}
)
resp.raise_for_status()
videos = resp.json().get("videos", [])
return [
{
"video_id": v.get("id"),
"description": v.get("description", ""),
"likes": v.get("likes", 0),
"shares": v.get("shares", 0),
"comments": v.get("comments", 0),
"views": v.get("views", 0),
"author": v.get("author", {}).get("username", ""),
"created": v.get("created_at", ""),
"engagement_rate": _calc_engagement(v)
}
for v in videos
]
def _calc_engagement(video: dict) -> float:
views = video.get("views", 1)
interactions = video.get("likes", 0) + video.get("shares", 0) + video.get("comments", 0)
return round(interactions / views * 100, 2) if views > 0 else 0.0步骤 3: 根据病毒潜力对产品进行评分
将主题标签增长、视频参与率和创作者多样性整合到一个病毒分数中。在多个创作者中得分较高的产品是最强烈的信号。
Python
def score_product(hashtag: dict, videos: list[dict]) -> dict:
unique_creators = len(set(v["author"] for v in videos if v["author"]))
avg_engagement = sum(v["engagement_rate"] for v in videos) / max(len(videos), 1)
total_views = hashtag.get("view_count", 0)
total_shares = sum(v["shares"] for v in videos)
# Viral score: weighted combination
score = (
(min(unique_creators, 10) / 10) * 30 + # creator diversity (30%)
(min(avg_engagement, 15) / 15) * 30 + # engagement rate (30%)
(min(total_shares, 5000) / 5000) * 20 + # share velocity (20%)
(min(total_views, 10_000_000) / 10_000_000) * 20 # reach (20%)
)
return {
"hashtag": hashtag.get("name", ""),
"viral_score": round(score, 1),
"unique_creators": unique_creators,
"avg_engagement_rate": round(avg_engagement, 2),
"total_views": total_views,
"total_shares": total_shares,
"top_video_description": videos[0]["description"][:200] if videos else "",
"analyzed_at": datetime.now().isoformat()
}步骤 4: 运行完整的管道并输出排名结果
扫描所有利基市场，对每个产品信号进行评分，并输出按病毒分数排序的获胜产品的排名列表。
Python
import asyncio
import json
async def find_winning_products() -> list[dict]:
all_products = []
for niche in PRODUCT_NICHES:
hashtags = await search_trending_hashtags(niche)
for hashtag in hashtags[:5]: # Top 5 per niche
videos = await fetch_hashtag_videos(hashtag.get("name", ""))
if videos:
product = score_product(hashtag, videos)
product["niche"] = niche
all_products.append(product)
# Sort by viral score
all_products.sort(key=lambda p: p["viral_score"], reverse=True)
return all_products
async def main():
products = await find_winning_products()
print(f"Analyzed {len(products)} product signals")
print(f"Credits used: ~{len(products) * 2} ({len(products) * 2 * 0.005:.2f})")
print("\nTop 5 Winning Products:")
for i, p in enumerate(products[:5], 1):
print(f" {i}. [{p['niche']}] #{p['hashtag']} - Score: {p['viral_score']}")
print(f" Creators: {p['unique_creators']} | Engagement: {p['avg_engagement_rate']}%")
print(json.dumps(products[:10], indent=2))
asyncio.run(main())Python 示例
Python
import asyncio
import httpx
SCAVIO_API_KEY = "your-api-key"
BASE_URL = "https://api.scavio.dev/api/v1/tiktok"
async def main():
async with httpx.AsyncClient(timeout=15) as client:
# Search hashtags
resp = await client.post(
f"{BASE_URL}/hashtag/search",
headers={"Authorization": f"Bearer {SCAVIO_API_KEY}"},
json={"query": "kitchen gadget", "limit": 10}
)
hashtags = resp.json().get("hashtags", [])
# Get videos for top hashtag
if hashtags:
top = hashtags[0]
resp = await client.post(
f"{BASE_URL}/video/search",
headers={"Authorization": f"Bearer {SCAVIO_API_KEY}"},
json={"query": top["name"], "limit": 5}
)
videos = resp.json().get("videos", [])
creators = len(set(v.get("author", {}).get("username", "") for v in videos))
print(f"Hashtag: #{top['name']}")
print(f"Views: {top.get('view_count', 0):,}")
print(f"Top videos: {len(videos)}")
print(f"Unique creators: {creators}")
asyncio.run(main())JavaScript 示例
JavaScript
const SCAVIO_API_KEY = "your-api-key";
const BASE_URL = "https://api.scavio.dev/api/v1/tiktok";
async function tiktokApi(endpoint, body) {
const resp = await fetch(BASE_URL + "/" + endpoint, {
method: "POST",
headers: { "Authorization": "Bearer " + SCAVIO_API_KEY, "Content-Type": "application/json" },
body: JSON.stringify(body)
});
return resp.json();
}
async function main() {
const hashtagData = await tiktokApi("hashtag/search", { query: "kitchen gadget", limit: 10 });
const hashtags = hashtagData.hashtags || [];
if (hashtags.length === 0) { console.log("No hashtags found"); return; }
const top = hashtags[0];
const videoData = await tiktokApi("video/search", { query: top.name, limit: 5 });
const videos = videoData.videos || [];
const creators = new Set(videos.map(v => v.author?.username).filter(Boolean)).size;
console.log("Hashtag: #" + top.name);
console.log("Views:", top.view_count?.toLocaleString());
console.log("Top videos:", videos.length);
console.log("Unique creators:", creators);
}
main();预期输出
JSON
Analyzed 30 product signals
Credits used: ~60 ($0.30)
Top 5 Winning Products:
1. [kitchen gadget] #miniwaffle - Score: 72.5
Creators: 8 | Engagement: 11.2%