Subgroup A: Recommendations for Enhancing Customer Behavior Insights and Sales Strategies

In today’s competitive e-commerce landscape, understanding and responding to customer behavior is critical for driving long-term growth and profitability. To effectively address key questions about customer purchasing behavior, retention, and the impact of marketing strategies, a comprehensive approach is essential. Identifying the factors that influence purchasing decisions allows businesses to predict and respond to customer needs, making targeted engagement possible. Retention efforts can be strengthened by calculating churn rates, identifying at-risk customers, and analyzing the effectiveness of retention strategies to maximize customer lifetime value (CLV). Additionally, determining the most effective marketing channels and campaigns enables companies to optimize their return on investment (ROI) and make data-driven decisions on promotional strategies. The following recommendations provide actionable insights into building loyalty, enhancing retention, and optimizing marketing strategies to foster sustained customer engagement and profitability.

---

1. Implement a Post-Campaign Retention Strategy with Personalized Touchpoints

• Objective: Extend customer engagement beyond campaign peaks, converting new customers from campaigns into loyal, returning buyers.
• Rationale: Our analysis shows that retention drops significantly after the initial months, indicating a lack of follow-up engagement. Personalized post-campaign communication can help retain customers by reinforcing their value to the brand.
• Action: Establish a multi-step, post-campaign engagement sequence, personalized to maintain the relationship:
  • Immediate Follow-Up: Within a week, send a personalized thank-you message along with tailored product recommendations. This immediate follow-up helps maintain interest and introduces customers to additional relevant products.
  • Two-Week Reminder: Offer an exclusive discount or incentive on related products based on past purchases. This nudge encourages a follow-up purchase while the campaign's impact is still fresh.
  • Monthly Loyalty Incentives: Provide a small reward, such as bonus points or free shipping, for any subsequent purchase within 30 days. This gesture rewards customers who continue to engage, creating a routine that can sustain between major campaigns.
• Expected Outcome: A structured, post-campaign retention approach can bridge the gap between campaigns, helping to stabilize CLV across months. This sequence of engagement not only keeps newly acquired customers active but also creates a steady cadence of interaction that smooths out non-promotional periods.

2. Data-Driven Personalization for Increased Relevance

• Objective: Maximize CLV by aligning engagement efforts with each customer’s specific purchase history and preferences.
• Rationale: Personalization based on data insights allows for timely, relevant engagement that enhances customer satisfaction and loyalty.
• Action:
  • Recommendation Algorithms: Use machine learning to develop product recommendations based on purchase history, browsing habits, and customer profiles with similar purchasing patterns. This approach increases the likelihood of conversions by presenting highly relevant product options.
  • Abandoned Cart and Browse Reminders: Send reminders for abandoned carts and popular products that customers have frequently viewed. Small incentives, such as a 5% discount, can encourage completion of these purchases.
  • Anniversary Offers: Use customer join dates and major purchase anniversaries to send personalized offers, enhancing emotional connection and making customers feel valued.
• Expected Outcome: Data-driven personalization deepens customer engagement by making each interaction meaningful and aligned with the customer’s needs. Tailored experiences foster loyalty, boosting both retention and CLV.

3. Subscription or Membership Program for Long-Term Engagement

• Objective: Establish consistent revenue and customer loyalty through a subscription-based model with regular exclusive perks.
• Rationale: Membership programs foster loyalty by providing consistent benefits, creating an additional revenue stream and encouraging long-term commitment.
• Action: Design a membership program that offers exclusive, ongoing benefits in exchange for a monthly or annual fee:
  • Exclusive Access: Members get priority access to limited-edition product lines or exclusive discounts, increasing perceived value and driving monthly engagement.
  • Regular Perks: Offer quarterly rewards, such as sample kits or store credit, to maintain customer interest and satisfaction.
  • VIP Support: Include enhanced customer service options, such as faster shipping and dedicated support for members, reinforcing the value of membership.
• Expected Outcome: A subscription program encourages repeat engagement and provides regular revenue. The exclusivity of membership benefits makes customers feel appreciated, leading to increased retention and stable CLV growth.

4. Predictive Analytics for Churn Prevention

• Objective: To prevent customer loss by identifying early signs of churn and re-engaging customers proactively.
• Rationale: Predictive analytics allow businesses to respond preemptively to churn indicators, reducing disengagement before customers fully lapse.
• Action:
  • Churn Risk Tiers: Segment customers into risk levels based on recent purchasing behavior, with a focus on high-risk customers.
  • Automated Re-Engagement: Send tailored discounts or loyalty points to high-risk segments, encouraging re-engagement with personalized offers.
  • Loyalty Bonuses for At-Risk Customers: Provide additional points, free shipping, or small rewards for customers who have not engaged in a certain period, offering a timely incentive to re-engage.
• Expected Outcome: Predictive analytics allow for early intervention, helping to retain at-risk customers and lower churn rates. A proactive engagement approach bolsters retention, contributing to a steady increase in CLV across all customer segments.

5. Maximize ROI with KOL and Social Media Investments

• Objective: Increase Shopee’s brand engagement and conversion rates by focusing on high-ROI marketing channels.
• Rationale: KOL and Social Media have demonstrated superior ROI and CTR, making them the most effective for broader brand reach and engagement.
• Action:
  • Increase Budget Allocation: Prioritize higher budgets for KOL and Social Media campaigns, especially around key promotional periods like Mega Sales.
  • Enhance KOL and Social Media Marketing: Expand advertisements to more Social Media platforms and launch monthly influencer-led campaigns on social media or co-create branded content with key influencers to promote high-value campaigns to strengthen customer loyalty.
• Expected Outcome: Higher budget allocation to these channels should amplify brand visibility and conversions, directly impacting sales performance and customer acquisition.

6. Enhance AOV through Targeted Promotions on Livestream Exclusives and Mega Sales

• Objective: Boost Average Order Value (AOV) by optimizing promotional strategies for high-value campaigns.
• Rationale: Livestream Exclusives and Mega Sales contribute to exceptionally high AOV, suggesting strong customer interest in these offerings.
• Action:
  • Optimize Livestream Content and Offers: Host exclusive livestream events with limited-time, high-value deals featuring top-selling products that encourage impulse buying. Use influencers or popular hosts to engage viewers and highlight product benefits in real-time, which increases trust and can drive higher sales.
  • Amplify Mega Sales with Premium Bundling Options: Design premium bundles specifically for Mega Sales events, combining complementary products that appeal to customers seeking value for larger purchases. Dynamic pricing or tiered discounting can be used within Mega Sales to encourage customers to add more items to their cart to reach the next discount level. Offering additional incentives like free shipping or loyalty points for higher spending tiers, making larger purchases more appealing.
  • Targeted SMS and Email Campaigns for Pre-Event Hype: Leverage SMS and email to build anticipation for Livestream Exclusives and Mega Sales. Send reminders leading up to these events, emphasizing limited-time offers or exclusive deals. Personalize communications based on customer preferences, recommending relevant high-value products or bundles to increase AOV.
• Expected Outcome: This strategy should drive larger transactions and maximize revenue per purchase, especially during campaign peaks for Shopee's top performing campaigns, further enhancing Shopee’s revenue.

7. Optimize Low-ROI Channels for Strategic, High-AOV Transactions

• Objective: Use SMS, Website, and Email for high-value, targeted promotions instead of routine marketing.
• Rationale: While these channels have a lower ROI, they contribute to high AOV for specific campaigns like Bundle Promotions and Livestream Exclusives.
• Action:
  • Selective Usage: Limit SMS and email to high-AOV campaigns for those channels such as Livestream Exclusives, Bundle Promotions, and Mega Sales.
  • Strategic Timing: Use these channels to deliver exclusive, time-sensitive offers, encouraging customers to make larger purchases.
  • Pre-Announce Key Events: Send email or SMS notifications ahead of livestreams or bundle offers to build anticipation.
• Expected Outcome: This approach could help to maximise the unique strengths of SMS, Website, and Email, driving high-value transactions without overextending the budget, potentially improving the ROI of these channels.

Subgroup B: Strategies for Effective Inventory Management and Optimal Pricing

Overview

To enhance our supply chain efficiency by reducing delivery times and minimizing return rates, we propose implementing two key strategies:

1. Return-to-Stock Optimization: Leverage returned products as local inventory to improve delivery times and customer satisfaction.
2. Automated 24/7 Chatbot: Reduce seller response times to common inquiries, providing immediate customer support and allowing sellers to focus on complex questions.

Return-to-Stock Optimization

1. Existing Infrastructure and Procedures

We currently have several systems and processes in place that efficiently manage returns and support customer satisfaction. Key components of our current infrastructure include:

• Return Processing: Our Singapore fulfilment center quickly manages returned products, providing a smooth experience for customers and supporting efficient restocking.
• Condition Assessment: A standardized checklist allows us to assess the condition and usability of returned items, ensuring that they meet quality standards for resale. This process ensures that only products in acceptable condition are offered as resale options.
• Return to Seller: Protocols are in place to return non-resalable items to overseas sellers after the condition assessment, including dispute resolution procedures to handle any discrepancies.

This infrastructure provides a solid foundation for expanding into local storage and optimized fulfillment of returned items.

2. Proposed Optimization

To reduce delivery times and optimize fulfillment, we propose a optimization strategy that repurposes returned products as local inventory in our Singapore fulfillment center. This approach allows us to fulfill new orders quickly and cost-effectively by utilizing locally available returned items. Overseas sellers will be able to store returned products at our Singapore fulfillment center.

This strategy will allow us to:

• Localize Inventory for Faster Fulfillment: Returned products in good condition are stored as local stock, allowing us to fulfill new orders directly from Singapore. This reduces reliance on international shipping, resulting in significantly shorter delivery times.
• Reduce Costs on Overseas Shipping: By repurposing returned items as local inventory, we can reduce costs associated with overseas shipping and customs fees, creating an additional financial benefit to offset storage expenses.

This storage approach is especially beneficial for high-demand and frequently returned products, as it allows us to quickly meet customer demand with locally available stock.

To maximize the benefit of local stock, we can also enhance the order fulfillment process with two specific features:

• Local Stock Badge: A “Local Stock” badge will be displayed on listings with returned items that are available for immediate domestic shipment. This badge will:
  • Attract Customers Seeking Fast Delivery: Customers looking for quick fulfillment will be encouraged to choose locally stocked items.
  • Increase Customer Trust: Local stock implies faster delivery and easier return processes, which can boost confidence in the purchase.
  • Promote Sustainability: By reducing the need for international shipments, this option supports our sustainability goals by lowering the carbon footprint.
• Streamlined Order Fulfillment Process: When a new order is placed, the system will automatically check for available returned items in the Singapore fulfillment center before sourcing from overseas. If a returned item is in stock locally:
  • It will be shipped directly to the customer, reducing delivery times and eliminating international shipping fees.
  • This automated process minimizes manual checks, ensuring efficient fulfillment and seamless integration with existing workflows.

3. Expected Outcomes

By integrating the storage and fulfillment processes into a single strategy, we can achieve:

• Faster Delivery Times: By fulfilling orders with local stock, we expect to reduce delivery times by an estimated 60-80% for these items, providing a significant improvement in customer experience.
• Reduced Return Rates: Faster delivery times are likely to reduce return rates. Currently, our data suggests that customers are more likely to return items if they experience long wait times. By minimizing these wait times, we can expect a drop in returns.
• Improved Customer Satisfaction and Retention: With faster deliveries and an improved fulfillment experience, customer satisfaction is expected to increase, leading to higher retention and loyalty rates.
• Cost Savings: By minimizing the need for overseas shipping and customs fees for locally stocked items, we anticipate considerable cost savings, which will offset the storage expenses and improve profitability.

4. Risk Assessment and Mitigation

We recognize that implementing these strategies presents potential risks. Below are the identified risks and our proposed mitigation plans:

• Storage Costs: Holding items in the local fulfillment center may incur storage costs.
  • Mitigation: Limit the storage period to one month and focus on high-demand items to minimize costs and maximize inventory turnover.
• Quality Concerns: Customers may be hesitant to purchase returned items.
  • Mitigation: Clearly label items with a quality guarantee and “like-new” condition tags, following our standard condition assessment.
• Demand Uncertainty: There may be insufficient demand for certain returned items stored locally, leading to unnecessary storage costs.
  • Mitigation: Prioritize storage for items with high turnover rates or predictable demand to ensure they are quickly sold.

Automated Chatbot to Reduce Seller Response Time

The high correlation between seller response time and return rate indicates that delays in seller responses contribute significantly to customer dissatisfaction, resulting in increased returns.

To address this, we propose implementing a Automated 24/7 Chatbot that assists sellers in providing timely, accurate responses to customer common inquiries, allowing sellers to focus on more complex queries. By streamlining the communication process for sellers, we aim to minimize return rates and improve the overall shopping experience.

The analysis of customer behavior reveals that longer seller response times are strongly associated with higher return rates. Many customers have routine questions about products—such as inquiries on fitting size, warranty, and return policies—that could be addressed instantly. Currently, sellers spend valuable time responding to these repetitive inquiries, which can delay responses to more complex questions and increase response times overall. This leads to customer dissatisfaction, frustration, and, consequently, a higher likelihood of returns.

1. Key Features of the Automated Chatbot

• Instant Responses for Common Questions:

  • Objective: To reduce response times by handling frequently asked questions automatically.
  • Functionality: The chatbot will be programmed to answer frequently asked questions instantly, including inquiries to common topics such as
    • Fitting Size: Offering guidance on sizing based on product-specific charts or customer feedback and fit recommendations.
    • Warranty Information: Offering details about warranty coverage and duration.
    • Return and Exchange Policies: Informing customers about return timelines, conditions, and procedures.
    • Customer Benefits: Customers receive immediate answers to their questions without waiting for seller responses, enhancing their satisfaction and confidence in purchasing.

• Seller-Preset Responses for Customization:

  • Objective: To provide sellers with flexibility and control over chatbot responses to ensure accuracy and relevance.
  • Functionality: Sellers can customize and preset answers to common questions, allowing the chatbot to provide tailored responses that reflect each seller’s specific policies and product details.
  • Seller Benefits: Sellers no longer need to respond to repetitive inquiries, allowing them to focus on complex questions that require personalized responses. This improves efficiency and reduces overall response times.

• Escalation for Complex Questions

  • Objective: To ensure that complex or unresolved questions receive appropriate attention from sellers.
  • Functionality: If the chatbot’s response doesn’t satisfy the customer or if the question is unique, customers will have the option to forward the inquiry to the seller directly. The chatbot will flag these messages as high-priority for the seller.
  • Customer Benefits: Customers can still receive personalized assistance when needed, ensuring that the automated system enhances—rather than replaces—the personal touch in customer service.

• Machine Learning for Continuous Improvement

  • Objective: To enhance the chatbot’s accuracy and relevance over time, making it more effective at handling complex customer inquiries.
  • Functionality: The chatbot will analyze patterns in customer questions and seller responses, learning from each interaction to improve future responses. Through machine learning, the chatbot can develop nuanced answers that address customer needs more effectively. Over time, this self-learning feature will reduce the number of questions that need to be escalated to sellers, further enhancing efficiency.
  • Long-Term Benefits: The system will become increasingly efficient and accurate, reducing the need for manual adjustments from sellers and providing an even better customer experience.

2. Expected Benefits

For Customers

• Immediate Support: Customers receive instant answers to their questions, reducing frustration, uncertaint and the likelihood of initiating returns caused by unaddressed questions or dissatisfaction with delayed responses.
• Improved Decision-Making: Timely responses about product details, sizing, and return policies help customers make informed purchasing decisions, reducing the likelihood of returns.
• Enhanced Experience: Customers benefit from instant answers, which improves their confidence in purchasing and provides a smoother pre-purchase experience. This could positively influence their decision to buy the product or to keep the product after purchase, reducing return rates.

For Sellers

• Reduced Response Time: By automating responses to common questions, improving the efficiency of their operations and reducing overall response times to meet response time targets.
• Efficiency Boost: The chatbot filters out routine inquiries, enabling sellers to focus on more high-priority, complex questions that require a personal response rather than spending time on repetitive inquiries. This enables them to provide high-quality responses where it matters most.
• Lower Return Rates: Faster and more accurate responses can help set correct expectations for the product and potentially reducing the likelihood of returns caused by unaddressed questions or dissatisfaction with delayed responses.
• Scalability: During peak periods, the chatbot can handle a high volume of inquiries without requiring additional seller resources.

For the Platform

• Higher Customer Satisfaction and Retention: A faster, more responsive support system improves the overall user experience, encouraging repeat purchases and loyalty.
• Reduced Operational Costs: Fewer escalated queries mean less strain on seller resources, allowing for more efficient use of personnel.
• Enhanced Data Collection: The chatbot can track common customer inquiries, providing insights into customer needs and potential product or service improvements.

3. Expected Outcomes

By implementing this automated chatbot solution, we anticipate the following outcomes:

• Reduced Response Times: Faster response times for common inquiries, improving customer satisfaction.
• Lower Return Rates: Enhanced customer decision-making reduces returns.
• Increased Seller Efficiency: Sellers can focus on complex inquiries, improving their operational efficiency and service quality.
• Continuous Improvement: Self-learning capabilities enable the chatbot to adapt and become more effective over time.