Product Analysis

Extract structured intelligence from any product listing — attributes, selling points, pricing insights, and market positioning.

Overview

The Product Analysis module uses multi-modal AI (text + vision) to understand products at depth. Feed it a product URL, title, images, or raw description — get back structured, actionable data.

Throughput: ~150ms per call | Batch: 10,000 products/hour

Basic Usage

from aizel_commerce import AizelCommerce

client = AizelCommerce()

result = client.product.analyze(
    title="Apple AirPods Pro (2nd Gen) with USB-C",
    description="Adaptive Audio. Personalized Spatial Audio. A new U2 chip...",
    images=[
        "https://example.com/airpods-front.jpg",
        "https://example.com/airpods-case.jpg"
    ],
    category="electronics/audio",
    market="amazon_us"
)

Response Schema

result.attributes        # Dict — extracted product attributes
result.selling_points    # List[str] — top selling points ranked by impact
result.pricing_insight   # PricingInsight — market positioning & price analysis
result.seo_keywords      # List[str] — high-value keywords
result.quality_score     # float (0-1) — listing quality assessment
result.suggestions       # List[str] — listing improvement recommendations
result.category_tags     # List[str] — marketplace category classification

Example Output

{
  "attributes": {
    "brand": "Apple",
    "type": "In-ear True Wireless",
    "connectivity": "Bluetooth 5.3",
    "battery_buds": "6h",
    "battery_case": "30h",
    "anc": true,
    "water_resistance": "IP54",
    "chip": "Apple H2 + U2",
    "weight": "5.3g per bud",
    "connector": "USB-C"
  },
  "selling_points": [
    "2x more effective Active Noise Cancellation vs previous gen",
    "Adaptive Audio automatically tunes sound to environment",
    "Personalized Spatial Audio with dynamic head tracking",
    "USB-C charging with precision finding via U2 chip",
    "IP54 dust and water resistance for workouts"
  ],
  "pricing_insight": {
    "market_position": "premium",
    "current_price": 249.00,
    "competitor_avg": 179.99,
    "suggested_range": [229, 249],
    "price_elasticity": 0.65,
    "discount_sensitivity": "low"
  },
  "quality_score": 0.87,
  "suggestions": [
    "Add comparison table vs AirPods Pro 1st gen",
    "Include battery life in title for search visibility",
    "Add lifestyle images showing workout usage"
  ]
}

API Reference

`POST /v1/product/analyze`

Parameter	Type	Required	Description
`title`	string	Yes	Product title
`description`	string	No	Full product description
`images`	array	No	Product image URLs (up to 10)
`category`	string	No	Category path (e.g., `electronics/audio`)
`market`	string	No	Target marketplace: `amazon_us`, `amazon_jp`, `shopify`, `taobao`, `rakuten`
`fields`	array	No	Specific fields to return (default: all)
`language`	string	No	Output language (default: `en`)

`POST /v1/product/batch_analyze`

Parameter	Type	Required	Description
`products`	array	Yes	Array of product objects
`fields`	array	No	Fields to extract
`concurrency`	int	No	Parallel workers (default: 10, max: 100)
`callback_url`	string	No	Webhook URL for async results

Batch Processing

import pandas as pd

catalog = pd.read_csv("catalog_10k.csv")

results = client.product.batch_analyze(
    products=catalog.to_dict("records"),
    fields=["selling_points", "pricing", "seo_keywords", "quality_score"],
    concurrency=50
)

# Process 10,000 products in ~60 minutes
print(f"Analyzed {len(results)} products")
print(f"Avg quality score: {results.avg('quality_score'):.2f}")

# Export results
results.to_csv("analyzed_catalog.csv")

On-Chain Verification

Enable cryptographic proof of analysis results:

result = client.product.analyze(
    title="...",
    on_chain=True,
    chain="arbitrum"
)

print(result.verification_hash)  # Content hash stored on-chain
print(result.tx_hash)            # Transaction receipt
# Useful for: supplier audits, authenticity claims, regulatory compliance

Use Cases

Catalog enrichment — Auto-fill missing attributes for thousands of SKUs
Listing optimization — Score and improve every product page
Market entry research — Analyze competitor products before launch
Price intelligence — Understand positioning across marketplaces
Content audit — Identify underperforming listings at scale

results matching ""

No results matching ""