E-commerce Apps & Catalog Systems: Design, Auctions & Recommenders

Key Concepts in E-Commerce Applications & Catalog Management

E-commerce applications enable businesses to sell products/services online, while catalog management organizes and displays products efficiently. This unit covers:

• Value chains and value webs in e-commerce
• Auction models (English, Dutch, reverse, etc.)
• Catalog design principles (hierarchical, faceted, semantic)
• Recommender systems (collaborative vs. content-based filtering)
• Omnichannel strategies (integration of online/offline sales)
• Security in e-commerce applications

1. Value Chain and Value Web in E-Commerce

Definition

A value chain is the sequence of activities a firm performs to deliver a product/service to the market. In e-commerce, it includes:

• Inbound logistics (supply chain management)
• Operations (product development, manufacturing)
• Outbound logistics (order fulfillment, delivery)
• Marketing & sales (digital ads, promotions)
• Service (customer support, returns)

A value web extends this by connecting multiple firms (suppliers, partners, customers) in a networked ecosystem.

Significance in E-commerce

• Efficiency: Automates logistics (e.g., Daraz’s warehouse automation).
• Customer-centricity: Personalizes marketing (e.g., NEPSE’s stock alerts).
• Collaboration: Enables B2B/B2C integration (e.g., Pathao’s ride-hailing + food delivery).

Comparison: Value Chain vs. Value Web

2. Auction Models in E-Commerce

Auctions are competitive bidding platforms where buyers/sellers interact dynamically.

Types of Auctions

flowchart TD
    A["Auction Types"] --> B["English Auction"]
    A --> C["Dutch Auction"]
    A --> D["Reverse Auction"]
    A --> E["First-Price Sealed Bid"]
    A --> F["Second-Price Sealed Bid"]
    B --> B1["Highest bidder wins\n(e.g., NEPSE stock auctions)"]
    C --> C1["Price starts high, drops until bidder accepts\n(e.g., eBay ‘Buy It Now’)"]
    D --> D1["Sellers bid for buyer’s business\n(e.g., government procurement)"]
    E --> E1["Bidders submit sealed bids, highest wins\n(e.g., Daraz seller auctions)"]
    F --> F1["Highest bidder pays second-highest price\n(e.g., Google AdWords)"]

How Auctions Work (Example: NEPSE Stock Auction)

1. Listing: Shares are listed with initial price.
2. Bidding: Investors submit bids (e.g., ₹100 for 100 shares).
3. Matching: Highest bids are matched with sellers.
4. Execution: Trades settle at the clearing price (e.g., ₹102 for 90 shares).

Advantages:

• Transparency (public bids).
• Efficiency (automated matching).

Disadvantages:

• Risk of manipulation (e.g., front-running).
• Complexity for new users.

3. Catalog Management in E-Commerce

A catalog organizes products for easy discovery. Key designs:

• Hierarchical: Categories → Subcategories → Products (e.g., Daraz’s "Electronics > Phones > Samsung").
• Faceted: Filters by attributes (price, brand, rating).
• Semantic: Uses AI to understand user intent (e.g., "best budget laptop" → filters accordingly).

Catalog Design Principles

1. User-Centric Layout:
   • IMAGE: e-commerce product catalog interface labelled diagram | Example: Daraz’s mobile app product grid with filters
   • Prioritize high-demand items (e.g., "Trending" section).
2. Search Optimization:
   • Autocomplete (e.g., typing "iPhone" suggests "iPhone 15 Pro").
   • Synonym handling (e.g., "smartwatch" = "fitness band").
3. Dynamic Updates:
   • Real-time stock alerts (e.g., "Only 3 left in stock!").

Worked Example: Daraz’s Catalog System

• Hierarchy: Electronics > Wearables > Smartwatches.
• Faceted Filters: Brand (Apple, Fitbit), Price (₹5,000–₹10,000), Rating (>4.5 stars).
• Personalization: "Recommended for you" based on past purchases.

4. Recommender Systems

Recommender systems suggest products/services based on user behavior.

Approaches

How Content-Based Filtering Works

1. Profile Creation: User’s past purchases (e.g., bought "Nike shoes").
2. Item Matching: System finds similar items (e.g., "Nike running shoes").
3. Recommendation: Displays top matches.

Example: Daraz’s "Frequently Bought Together"

• If you buy a laptop, it suggests a mouse/keyboard (based on purchase data).

How Collaborative Filtering Works

1. User-Item Matrix:
2. Similarity Calculation: Finds users like Alice (e.g., Bob) and recommends their items.
3. Prediction: "Bob also bought headphones → Recommend to Alice."

5. Omnichannel E-Commerce

Omnichannel integrates online and offline sales for a seamless experience.

Key Elements

• Consistency: Same pricing/promotions across channels (e.g., Daraz’s app and website).
• Data Sharing: Purchase history syncs (e.g., Pathao’s app remembers your ride preferences).
• Unified Inventory: Real-time stock updates (e.g., NEPSE’s stock market app).

Example: NEPSE’s Omnichannel Presence

• Online: Trading via app/website.
• Offline: Brokerage firms (e.g., Global IME, Saubhagya) assist users.
• Integration: Mobile app shows real-time broker recommendations.

6. Security in E-Commerce Applications

Security threats include:

• Phishing: Fake login pages (e.g., "eSewa support" scams).
• Payment Fraud: Stolen credit cards (e.g., unauthorized Daraz purchases).
• Data Breaches: Hacked customer databases (e.g., Ncell’s SIM swap attacks).

Security Measures

In the Real World

1. Daraz’s Auction System:

   • Uses reverse auctions for sellers to bid for prime placement in search results (higher bids = better visibility).
   • Recommender systems suggest products based on browsing history (e.g., "You viewed a phone → see accessories").

2. NEPSE’s Stock Auctions:

   • English auction for buying/selling shares in real-time.
   • Value web: Connects brokers, investors, and clearing houses (e.g., Nepal Stock Exchange Limited).

3. Pathao’s Omnichannel Logistics:

   • Value chain: Rider pickup → GPS tracking → delivery confirmation.
   • Catalog integration: Users can order food via Pathao’s app while booking a ride.

Exam Tip

• Focus on comparisons: Value chain vs. web, auction types, recommender methods.
• Apply to real examples: Always tie theory to Daraz, NEPSE, or Pathao.
• Diagrams are key: Draw flowcharts for auctions, catalog hierarchies, or recommender systems.
• Security questions: Expect 1–2 marks on threats (phishing, fraud) and solutions (MFA, encryption).
• Omnichannel: Explain how integration (e.g., NEPSE’s app + brokers) improves user experience.

Sample Exam Answer Structure:

1. Define (e.g., "A value web is a network of interconnected firms...").
2. Explain (e.g., "Daraz connects with Ncell for logistics...").
3. Compare (e.g., "Unlike a value chain, a web is dynamic...").
4. Example (e.g., "NEPSE’s stock auctions use English bidding...").