K-Means Dominates E-commerce Segmentation Despite Evolving Data

Why It Matters

Understanding how customers can be grouped is fundamental for effective marketing and product development. This insight highlights a common, albeit potentially outdated, approach used in practice, suggesting opportunities for more advanced or tailored segmentation strategies.

Key Finding

The research identified a standard four-step process for customer segmentation in e-commerce and found that k-means clustering is the most common technique used, despite the increasing complexity of data.

Key Findings

• The four-phase process of customer segmentation involves information collection, customer representation, segmentation analysis, and customer targeting.
• K-means is the most frequently used segmentation method across various e-commerce use cases and data sizes.
• Customer representation is often achieved through manual feature selection or RFM analysis.

Research Evidence

Aim: What are the predominant methods for customer segmentation in e-commerce, and how have they evolved over time?

Method: Literature Review

Procedure: A comprehensive literature search was conducted, identifying 105 publications from 2000-2022 related to customer segmentation in e-commerce. The identified methods were analyzed for temporal trends and applicability to different data dimensionalities, leading to a four-phase process model.

Context: E-commerce customer targeting and marketing

Design Principle

Leverage data-driven segmentation to inform personalized user experiences and targeted marketing efforts.

How to Apply

When developing a new e-commerce platform or marketing campaign, analyze your target audience using segmentation methods, and consider if k-means is sufficient or if more advanced techniques are needed.

Limitations

The review focuses on published literature and may not capture all proprietary methods used in industry. The dominance of k-means might also reflect its ease of implementation rather than its optimal performance for all scenarios.

Student Guide (IB Design Technology)

Simple Explanation: Most online shops group their customers using a method called k-means, even though customer data is getting more complicated.

Why This Matters: Understanding how to group users helps you design products and experiences that better meet the needs of different customer segments.

Critical Thinking: Given the dominance of k-means, what are the potential drawbacks of this method for creating truly personalized user experiences, and what alternative approaches could yield more nuanced customer insights?

IA-Ready Paragraph: Research indicates that k-means is a widely adopted method for customer segmentation in e-commerce, often used in conjunction with manual feature selection or RFM analysis for customer representation. While prevalent, the continued reliance on k-means warrants consideration of its suitability for increasingly complex and high-dimensional datasets, suggesting potential opportunities for exploring more advanced segmentation strategies in design projects.

Project Tips

• When defining your target audience, consider how you will segment them.
• Research different segmentation algorithms beyond k-means to see if they offer better insights for your specific design problem.

How to Use in IA

• Use this research to justify your choice of segmentation method or to discuss the limitations of commonly used methods in your design project.

Examiner Tips

• Demonstrate an awareness of the limitations of common segmentation techniques and explore more advanced methods if appropriate for your design context.

Independent Variable: Segmentation methods (e.g., k-means, manual feature selection, RFM analysis)

Dependent Variable: Effectiveness of customer targeting, customer representation quality, applicability to data dimensionality

Controlled Variables: E-commerce use cases, time period of publications (2000-2022)

Strengths

• Comprehensive literature review covering a significant time span.
• Analysis of temporal trends and data dimensionality applicability.

Critical Questions

• Are there emerging segmentation methods that are not yet widely published but are being used in industry?
• How does the choice of segmentation method impact the design of user interfaces and marketing communications?

Extended Essay Application

• Investigate the impact of different segmentation strategies on user engagement metrics for a specific e-commerce product or service.

Source

A review on customer segmentation methods for personalized customer targeting in e-commerce use cases · Information Systems and e-Business Management · 2023 · 10.1007/s10257-023-00640-4