Skip to content

harshil0014/SmartCart_Clustering_system

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 

Repository files navigation

E-commerce Customer Segmentation System

This project identified four distinct customer segments based on income, spending, and engagement. High-value customers showed low browsing but high purchases, while budget segments required promotional strategies. This enables targeted marketing and improved business decision-making.

                CUSTOMER SEGMENTATION

    ┌──────────────────────┬────────────────────────┐
    │   C0: FAMILY         │   C1: LOYALTY PROGRAM  │
    │   SHOPPERS           │                        │
    ├──────────────────────┼────────────────────────┤
    │ • More children      │ • Fewer children       │
    │ • Poor response      │ • Slightly older       │
    │ • With partners      │ • Avg response         │
    │ • High web visits    │ • With partners        │
    │ • Low purchases      │ • Store/Catalog ↑      │
    │                      │ • Web ↓                │
    │ 🎯 Strategy:         │ 🎯 Strategy:           │
    │ Discounts, bundles   │ Loyalty programs       │
    └──────────────────────┴────────────────────────┘

    ┌──────────────────────┬────────────────────────┐
    │   C2: DIGITAL        │   C3: HIGH VALUE        │
    │   BROWSERS           │   CUSTOMERS (BEST ROI)  │
    ├──────────────────────┼────────────────────────┤
    │ • More children      │ • Fewer children       │
    │ • Avg response       │ • Slightly older       │
    │ • Mostly alone       │ • Best response        │
    │ • High web visits    │ • Mostly alone         │
    │ • Very low purchases │ • Store/Catalog ↑↑     │
    │                      │ • Web ↓                │
    │ 🎯 Strategy:         │ 🎯 Strategy:           │
    │ Heavy discounts      │ Premium services       │
    │ Conversion focus     │ High-value targeting   │
    └──────────────────────┴────────────────────────┘

alt text

🔵 Cluster 0 → Low-Spend Families

i) Income: ~39k (low-mid) ii) Spending: ~222 ❌ low iii) Purchases: low iv) Web visits: high (~6.3) v) Children: high (~1.24)

👉 Profile:

  1. Budget-conscious families
  2. Browse a lot but don’t buy much

👉 Business strategy:

  1. Discounts / offers
  2. Family bundles

🔴 Cluster 1 → High-Value Premium Customers

i) Income: ~72k ✅ high ii) Spending: ~1236 🚀 very high iii) Purchases: high iv) Web visits: low (~3.5) v) Children: low (~0.5)

👉 Profile:

  1. Rich, decisive buyers
  2. Don’t browse much → already know what they want

👉 Business strategy:

Premium products:

  1. Loyalty rewards
  2. Upselling

🟡 Cluster 2 → Low Income – Low Engagement

i) Income: ~36k ❌ lowest ii) Spending: ~165 ❌ lowest iii) Purchases: low iv) Web visits: high (~6.6)

👉 Profile:

  1. Price-sensitive
  2. High browsing, low conversion

👉 Business strategy:

  1. Heavy discounts
  2. Entry-level products

🟢 Cluster 3 → High Income – Opportunistic Buyers

i) Income: ~70k ✅ high ii) Spending: ~1190 🚀 high iii) Purchases: high iv) Web visits: moderate v) Response: 0.32 (highest!) 🔥

👉 Profile:

  1. Respond well to campaigns
  2. High potential growth segment

👉 Business strategy:

  1. Targeted marketing campaigns
  2. Personalized offers

About

Built a customer segmentation model using KMeans and Agglomerative Clustering, identifying 4 distinct customer groups and deriving actionable marketing strategies based on income, spending behavior, and engagement patterns.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors