Cross-Platform SKU Mapping and Unified Metric System: Building a Standardized View of Equivalent Products Across E-Commerce Sites

Core Objectives

  • Build a cross-platform product database to associate identical products across multiple e-commerce platforms

  • Standardize key metrics such as price, inventory, and sales volume into a unified KPI pool

  • Develop SKU-level monitoring dashboards with real-time alerting (e.g., sudden price increases or stockouts)

Step 1: Collect Basic Product Data Across Platforms

Using Lazada, Pinduoduo, and Amazon as examples, we fetch product details via the LuckData API to prepare for matching and data consolidation.

Lazada Product Data Retrieval

import requests

def get_lazada_product_detail(site, item_id):

url = "https://luckdata.io/api/lazada-online-api/x3fmgkg9arn3"

params = {

"site": site, # Supports "vn", "th", "ph"

"itemId": item_id

}

res = requests.get(url, params=params)

return res.json()

lazada_data = get_lazada_product_detail("vn", "2396338609")

print(lazada_data["data"]["title"], lazada_data["data"]["price"])

Pinduoduo Product Data (Simulated)

Data can be obtained via custom web scrapers or LuckData’s Pinduoduo interface.

pdd_data = {

"title": "Bear Electric Lunch Box, Double Layer",

"price": 129.0,

"sku_id": "pdd_948571",

"image": "https://cdn.example.com/pdd.jpg"

}

Amazon Product Data

amazon_data = {

"title": "Bear Electric Lunch Box, 2-Tier Food Steamer",

"price": 34.99,

"asin": "B09XY1234L",

"image": "https://cdn.example.com/amazon.jpg"

}

Core Algorithm: Matching and Aggregating Identical SKUs

✅ Method 1: Title Similarity Matching

Use FuzzyWuzzy or RapidFuzz to determine if product titles indicate the same item.

from rapidfuzz import fuzz

def is_same_product(title_a, title_b, threshold=80):

score = fuzz.token_sort_ratio(title_a.lower(), title_b.lower())

return score > threshold

matched = is_same_product(lazada_data["data"]["title"], amazon_data["title"])

print("Same product:", matched)

Weighted scoring is recommended:

  • Title similarity (70%)

  • Image hash similarity (15%)

  • Brand/model similarity (15%)

✅ Method 2: Standardized SKU Schema

Create a unified SKU ID for each logically identical product and map corresponding entries from each platform:

{

"sku_id": "SKU_001",

"standard_title": "Bear Electric Lunch Box 2-Tier",

"platforms": {

"lazada_vn": {"item_id": "2396338609", "price": 135000, "url": "..."},

"pinduoduo": {"sku_id": "pdd_948571", "price": 129.0},

"amazon": {"asin": "B09XY1234L", "price": 34.99}

}

}

This serves as the data model for metrics aggregation and dashboard construction.

Unifying Metrics: Price, Inventory, and Sales

Build a Standardized Daily Metric Table

SKU ID

Platform

Product Title

Price

Inventory

Sales

Date

SKU_001

Lazada_vn

Bear Electric Lunch Box

135000

54

320

2025-05-21

SKU_001

Pinduoduo

Bear Electric Lunch Box (CN)

129.0

68

480

2025-05-21

SKU_001

Amazon

Bear Electric Lunch Box (EN)

34.99

23

890

2025-05-21

Sample Dashboard Display Options

Tools you can use:

  • Streamlit + Pandas: Lightweight web-based dashboards

  • Google Data Studio: Integrate with Sheets for fast deployment

  • PowerBI / Tableau: Enterprise-grade visual analytics

Alerting and Smart Monitoring

✅ Example: Price Fluctuation Alert

Monitor for abnormal price changes beyond a defined threshold (e.g., 15%) and trigger an alert.

def price_alert(sku_id, price_today, price_yesterday):

delta = abs(price_today - price_yesterday) / price_yesterday

if delta > 0.15:

return f"[Alert] SKU {sku_id} price fluctuated over 15%"

Use scheduled tasks (e.g., Airflow / CRON) to automate monitoring and push alerts to channels like Slack or Lark.

Future Roadmap: Enhancing Matching Capabilities

Stage

Focus Area

V1

Title similarity + manual SKU mapping

V2

Image hash comparison + rule-based parsing

V3

AI model for "image + title" product matching and clustering

The evolution moves from simple title comparison to multimodal AI-based identification of identical products.

✅ Summary

  • Use APIs to quickly build multi-platform product datasets for Lazada, Pinduoduo, and Amazon

  • Apply similarity metrics to construct a unified SKU repository

  • Consolidate metrics like price, inventory, and sales by SKU

  • Enable price monitoring, competitor comparison, and real-time alerts

  • Lay the groundwork for intelligent, cross-platform product operations and analysis

Articles related to APIs :