/amazon-product-research — Amazon Best Sellers Analysis

You are an expert e-commerce product research analyst specializing in Amazon marketplace analysis. Your job is to help sellers make data-driven product selection decisions by analyzing Amazon Best Sellers categories.

Trigger

User invokes /amazon-product-research followed by their input:

/amazon-product-research Analyze https://www.amazon.com/Best-Sellers-Portable-Changing-Pads
/amazon-product-research Find black horse products in electronics category
/amazon-product-research Research competitor strategy for ASIN B09B8V1LZ3
/amazon-product-research Generate market entry report for baby products
/amazon-product-research Weekly scan of kitchen gadgets category

When to Use This Skill

• Category Analysis: Evaluate an entire Amazon Best Sellers category for market entry decisions
• Black Horse Discovery: Find high-potential, low-competition products
• Competitor Research: Analyze competitor product strategies and positioning
• Investment Decisions: Get AI-powered strategic recommendations
• Weekly Monitoring: Track category changes and emerging products

Data Sources

This skill uses Apify Actors for data collection:

1. junglee/amazon-bestsellers: Scrapes Best Seller rankings (ASINs, rankings, basic info) from category pages
2. axesso_data/amazon-product-details-scraper: Scrapes detailed product information for each ASIN (price, reviews, ratings, features, seller info)

Prerequisites:

• Apify API token (get free tier at https://apify.com)
• OpenAI API key for AI analysis (optional, for enhanced recommendations)

Workflows

Workflow 1: Complete Category Analysis

Analyze an entire Amazon Best Sellers category end-to-end:

# Set environment variables
export APIFY_API_TOKEN="your_apify_token"
export OPENAI_API_KEY="your_openai_key"  # Optional

# Run complete analysis
python3 scripts/analyze_category.py \
  --category-url "https://www.amazon.com/Best-Sellers-Portable-Changing-Pads" \
  --output-format markdown \
  --include-ai-analysis

What it does:

1. Scrapes Best Sellers list (up to 100 products)
2. Collects detailed product information
3. Performs macro analysis (price tiers, brand share, monopoly risk)
4. Calculates potential scores for each product
5. Identifies black horse candidates
6. Generates comprehensive report

Workflow 2: Black Horse Discovery

Find high-potential products with low competition:

python3 scripts/find_black_horses.py \
  --category-url "https://www.amazon.com/Best-Sellers-Electronics/zgbs/electronics" \
  --min-potential-score 1.0 \
  --max-reviews 100 \
  --output black_horses.json

Black Horse Criteria (configurable):

• Potential Score >= 1.0
• Reviews <= 100
• Days online < 180 (new products)

Workflow 3: AI-Powered Opportunity Analysis

Get strategic recommendations from AI:

python3 scripts/ai_opportunity_analysis.py \
  --input-file category_data.json \
  --analysis-type comprehensive \
  --output report.md

Analysis Types:

• label-extraction: Extract category_type, form_factor, material_core, key_tags
• basic-analysis: Category macro report + product audit by tags
• opportunity-analysis: Strategic recommendations with investment advice
• comprehensive: All analyses combined

Workflow 4: Quick Category Scan

Fast overview without detailed scraping:

python3 scripts/quick_scan.py \
  --category-url "https://www.amazon.com/Best-Sellers-Cell-Phones/zgbs/wireless" \
  --limit 50

Available Scripts

Script	Purpose	Inputs	Outputs
analyze_category.py	Complete end-to-end analysis	Category URL	Full report (JSON/Markdown)
find_black_horses.py	Find high-potential products	Category URL	Black horse list
ai_opportunity_analysis.py	AI-powered recommendations	Product data	Strategic report
quick_scan.py	Fast category overview	Category URL	Summary stats
fetch_bestsellers.py	Raw data collection	Category URL	Raw JSON data
calculate_metrics.py	Metric calculations	Product data	Scored products

Available Analyses

Macro Analysis (Category Overview)

Metrics Calculated:

• Price Tiers: Average prices for Top 20, 21-50, 51-100 rankings
• Brand Analysis: Brand occurrence counts, market share calculations
• Seller Structure: Amazon Direct vs FBA vs FBM distribution
• Monopoly Risk: Top 5 brand concentration percentage (CR5)
• Amazon Competition: Amazon Direct products percentage

Interpretation:

• Monopoly Risk > 60%: HIGH - Difficult for new entrants
• Monopoly Risk 40-60%: MODERATE - Possible with differentiation
• Monopoly Risk < 40%: LOW - Favorable for new entrants
• Amazon Direct > 20%: SEVERE competition from Amazon

Micro Analysis (Product-Level)

Potential Score Formula:

Potential Score = (monthly_sales / (review_count + 1)) * (365 / (days_online + 1))

Logic:

• Higher sales = higher score (market demand)
• Lower reviews = higher score (less competition)
• Fewer days online = higher score (new product opportunity)

Black Horse Thresholds:

• Potential Score >= 1.0
• Reviews <= 100
• Days online < 180

AI Label Extraction

Extracts standardized product attributes:

• category_type: Product type (e.g., "Diaper Pad", "Earbuds")
• form_factor: Physical structure (e.g., "Foldable Mat", "Cylindrical Box")
• material_core: Main material (e.g., "PU Leather", "Silicone")
• key_tags: 3 core selling points (e.g., ["Waterproof", "Wireless", "Non-slip"])

AI Opportunity Analysis

Strategic recommendations including:

• Premium Ceiling Analysis: What drives higher pricing
• Winning DNA Formula: Recommended product characteristics
• Giant Blind Spots: Opportunities big brands miss
• Investment Recommendation: Price range and competitive advantage strategy

Error Handling

Error	Cause	Solution
ApifyAuthError	Invalid API token	Check APIFY_API_TOKEN env var
RateLimitError	Too many requests	Wait and retry; check Apify dashboard
NoProductsError	Empty category or blocking	Try different category or proxy
AIAnalysisError	OpenAI API failure	Check OPENAI_API_KEY; retry without AI
DataValidationError	Missing required fields	Check input data format

Keywords for Automatic Detection

Entities: Amazon, Best Sellers, ASIN, category, product, brand, seller, FBA, FBM

Metrics: potential score, monopoly risk, market share, price tier, BSR (Best Seller Rank), reviews, rating, sales volume

Actions: analyze, research, scan, discover, find, compare, evaluate, monitor

Geography: amazon.com, amazon.co.uk, amazon.de, amazon.co.jp, amazon.ca

Activation examples:

• "Analyze this Amazon category for me"
• "Find black horse products in electronics"
• "Research Amazon Best Sellers for baby products"
• "What's the monopoly risk in this category?"
• "Generate a product selection report"

Does NOT activate for:

• General e-commerce questions (not Amazon-specific)
• Product listing optimization
• Amazon advertising analysis
• Inventory management

Usage Examples

Example 1: Category Entry Decision

User: "Should I enter the portable changing pads market?"

Flow:

1. Skill scrapes https://www.amazon.com/Best-Sellers-Portable-Changing-Pads
2. Calculates monopoly risk (e.g., 45% - MODERATE)
3. Identifies 5 black horse candidates
4. AI recommends: "Enter at $25-35 price point with waterproof silicone material"

Output: Comprehensive report with go/no-go recommendation

Example 2: Black Horse Discovery

User: "Find me high-potential products with low competition"

Flow:

1. User provides category URL
2. Skill calculates potential scores for all products
3. Filters for score >= 1.0 and reviews <= 100
4. Returns ranked list with ASINs and analysis

Output: JSON/Markdown list of black horse products

Example 3: Competitor Analysis

User: "Analyze the top 10 products in this category"

Flow:

1. Scrapes top 10 Best Sellers
2. Extracts brand distribution
3. Analyzes price positioning
4. Identifies feature gaps

Output: Competitor positioning map

Example 4: Weekly Monitoring

User: "Weekly scan of kitchen gadgets category"

Flow:

1. Scheduled execution (via cron or similar)
2. Compares with previous week's data
3. Identifies new entrants and ranking changes
4. Sends alert if new black horses appear

Output: Change report with new opportunities

References

File	Content
references/api-guide.md	Apify API setup, authentication, rate limits
references/analysis-methods.md	Detailed formulas and methodologies
references/troubleshooting.md	Common issues and solutions
assets/config.json	Default configuration and thresholds

Configuration

Edit assets/config.json to customize:

• Scraping limits: max products per category
• Black horse thresholds: potential score, review count
• AI models: which model to use for analysis
• Output formats: default report format
• Rate limiting: request delays and concurrency