DEV Community

kk mors
kk mors

Posted on

I Built an AI Product Trend Radar That Spots Winning Products Before Competitors

Every day, thousands of products trend on Amazon and TikTok Shop. By the time most sellers notice, the early-mover advantage is gone.

I built a Product Trend Radar — an AI-powered commerce intelligence system that scans Amazon and TikTok Shop daily to find trending products, analyze competition, mine review insights, and score opportunities before they saturate.

How It Works

The radar runs on 6 AI agent skills working together:

1. Amazon Product Scout

Scans bestsellers, new releases, and movers and shakers across your target categories. Extracts ASIN, price, rating, review count, BSR, and seller information.

2. TikTok Shop Scout

Monitors trending products, viral content, and creator partnerships. Measures content velocity — the number of videos, views, and engagement a product generates in 24h/7d windows.

3. Review Miner

Deep-mines product reviews to find:

  • What customers love (copy these features)
  • What they hate (fix these pain points)
  • What they wish existed (build these features)
  • Which competitors they mention

4. Competitor Tracker

Monitors competitor pricing, listing changes, inventory signals, and ranking movements. Alerts you when a competitor drops price, adds new variations, or runs out of stock.

5. Opportunity Scorer

Multi-factor scoring engine that rates each product 0-100 across 5 dimensions:

Dimension Weight What It Measures
Demand Signal 25% BSR trend, search volume, TikTok views
Competition Level 20% Seller count, brand concentration
Review Sentiment 20% Pain points, feature gaps
Content Velocity 15% Social momentum, creator activity
Profit Potential 20% Margins, shipping, seasonality

80+ = Green light. 60-79 = Investigate. Below 60 = Skip.

6. Daily Briefing Generator

Combines all signals into a concise morning report with:

  • Market pulse (biggest movers)
  • Hot opportunities with scores and margins
  • Competitor alerts
  • Prioritized action items

The Scoring Framework in Practice

# Example: Scoring a stackable bento box
product = {
    "title": "Stackable Bento Box",
    "bsr": 1923,           # Improving rapidly
    "reviews": 156,         # Low competition (few reviews)
    "rating": 3.8,          # Room for improvement
    "tiktok_views": 500000, # Viral content
    "seller_count": 5,      # Fragmented market
    "estimated_margin": 68, # Great margins
}

# Result: 84/100 — High Opportunity
# Action: Source immediately from 1688 at $2.10, sell at $19.99
Enter fullscreen mode Exit fullscreen mode

What You Get

The Product Trend Radar Pack includes:

  • 6 AI agent skill files — Ready-to-use prompts for any AI agent (Hermes, OpenClaw, Manus, ChatGPT)
  • 3 Python scripts — Amazon stealth scraper, opportunity scoring engine, configurable settings
  • Templates — Daily briefing, product scorecard, competitor profile
  • Examples — Sample outputs for every component

Works with or without coding. The AI agent skills need zero code — just copy the prompt and run.

Target Markets

  • Amazon US, EU (DE/UK/FR/IT/ES), Japan
  • TikTok Shop (Global)
  • Wildberries / Ozon (Russia)
  • Mercado Libre (LatAm)
  • 1688 (sourcing)

My Results

After running this radar for 3 months:

  • Found 12 products scoring 80+ before they peaked
  • Average time from detection to listing: 5 days
  • 3 products became top-10 BSR in their categories
  • One product generated $8K in its first month

The key insight: speed beats scale. Finding a trending product 2 weeks early is worth more than having 100 mediocre listings.

If you sell on Amazon, TikTok Shop, or any e-commerce platform, the Product Trend Radar Pack will help you spot winners before the crowd.

What tools do you use for product research? Would love to compare notes.

Top comments (0)