Practical Guide to Predicting Bestsellers: Using TikTok and Douyin Data to Spot the Next E-Commerce Trend

Introduction

"A breeze can stir from the smallest ripple." A seemingly ordinary short video can often signal the rise of the next bestselling product.

Today’s e-commerce marketing is no longer about following sales trends—it's about anticipating them based on traffic signals and preparing stock ahead of time.

This article takes a practical approach to show how real-time data from TikTok and Douyin—such as trending topics, interaction trends, and challenge propagation—can be leveraged to predict potential bestsellers. By cross-verifying with e-commerce platform data from Amazon, Lazada, and others, sellers can build a reliable, data-driven product selection strategy.

1. What is “Short Video Data-Driven Product Selection”?

Traditional product selection relies on experience, industry reports, and competitor sales analysis. Now, platforms like TikTok and Douyin provide a new front-end data stream: social video content and user engagement trends.

Key behavioral signals include:

  • A viral challenge → surge in sales of related product categories

  • An influencer’s video gains rapid traction → click-through and conversion rates rise for associated products

  • Sudden spike in topic hashtags → increased visibility and consumer curiosity (“seeding” effect)

These indicators serve as early signals of future demand. Recognizing and quantifying these patterns allows merchants to identify winning products early.

2. What Key Data Should We Capture?

Using LuckData’s APIs for TikTok and Douyin, sellers can easily access the following critical data dimensions:

Douyin Key APIs:

Function

Description

City Trending List

/get_xv5p: Daily trending videos by city, including play count and like growth

Video Trend Detail

/get_pa29: Track video topics, creators, and popularity over time

TikTok Key APIs:

Function

Description

Challenge Info

search challenge + challenge info: Track total plays, participating videos, and growth rates

Video Detail

tiktok video info: Monitor likes and comment trends to identify breakout points

✅ All APIs support filters by date, keyword, and region for targeted analysis.

3. Practical Workflow: How to Predict Potential Bestsellers

To identify promising products from video trends, use the following structured approach:

Step 1: Schedule Daily Retrieval of Douyin and TikTok Trending & Challenge Lists

# Douyin City Trending List Example

url = "https://luckdata.io/api/douyin-API/get_xv5p?city=110000&type=rise_heat&start_date=20240512&end_date=20240513"

Key fields include: video ID, title, creator, publish time, play count, like count, comment count, topic tags, etc.

Step 2: Build Keyword Cloud and Popularity Scoring Model

  • Use NLP to extract high-frequency keywords from video titles and challenge names

  • Apply a weighted scoring formula:
    play count × 0.5 + like count × 0.3 + comment growth × 0.2

  • Generate daily/weekly hot keyword rankings to detect emerging product concepts

This allows identification of spikes in interest for terms like “exfoliating gel,” “sunburn repair,” or “cooling balm.”

Step 3: Validate Keywords Against E-Commerce Platforms

Use product-related keywords to query APIs from Amazon, Lazada, or Shopee and assess real-world sales impact.

# Search for “exfoliating gel” on Lazada to see if products are gaining traction

url = "https://luckdata.io/api/lazada-online-api/gvqvkzpb7xzb?page=1&site=ph&query=去角質凝膠"

Validation criteria:

  • Product listings exist and show recent spikes in reviews or pricing volatility → likely gaining momentum

  • No listings yet → potential opportunity for sourcing/importing or early inventory placement

4. Case Study: From Douyin Video to Sales Spike on Lazada

Example: Douyin Challenge #ClayMaskCleaningTest

  • Video views exceeded 8 million

  • Likes surpassed 600,000

  • Hashtag popularity increased by 240% in 3 days

Searched “clay mask” on Lazada Malaysia:

  • Review count jumped from 120 to 680 within days

  • Price increased from RM22–28 to RM29–35, suggesting high demand

  • Product tagged as "Hot Pick" and promoted on the homepage

This case illustrates how a trend can transition from social video to e-commerce sales in less than a week. Through automated API queries and lightweight analysis, the entire discovery and validation process can be executed in under an hour—compared to traditional 2–3 week research cycles.

5. Advantages of Using LuckData API

Metric

Traditional Approach

LuckData API Approach

Data Coverage

Manual collection, inconsistent sources

Integrated data from TikTok, Douyin, Lazada, Amazon

Update Speed

Difficult to maintain daily updates

Real-time and scheduled data access

Technical Barrier

Requires custom crawlers, site parsing

Ready-to-use structured APIs

Cost & Efficiency

High labor and tech overhead

Free trial available, scalable to demand

✅ Result: Streamlined, accurate, and scalable trend-to-product pipeline

6. Conclusion: The Future of Product Selection Starts with Trend Insight

The gap between “what consumers are buying” and “what consumers are about to want” is simply a matter of data latency.

By tapping into the trend signals from TikTok and Douyin, sellers can identify demand before it materializes. Coupled with cross-verification on e-commerce platforms, product selection becomes a data-driven, actionable process—not a guessing game.

In a world where short-form video drives consumer behavior, every trending clip or viral challenge may hide the next multimillion-selling opportunity. All you need is the right insight—and the tools to act on it.

Articles related to APIs :