Reference no: EM134042457

Customer Analytics and Social Media

Assignment

Customer Segmentation and Profiling

Learning Objective

The objective of Assignment 2 is to develop your understanding and practical skills in customer segmentation and customer retention analytics. Using a banking churn case study, you will apply clustering and profiling techniques to identify meaningful customer segments, evaluate how churn risk differs across segments, and generate actionable recommendations for customer retention strategy and targeting.

Assignment 2 is aimed to develop the customer analytics skills via performing customer segmentation and profiling tasks based on a case study. This assessment will contribute 35% to your overall grade of this unit.

Details:
Customer segmentation is a pivotal task for business analytics. Customer segmentation is the process of splitting customers into different groups with similar characteristics for potential business value proposition. Many companies find that segmenting their customers enable them to communicate, engage with their customers more effectively.

A bank is conducting an analysis on the existing customer profiles and the marketing campaign data to identify the target customers who are mostly likely to subscribe long-term deposits. As a member of the data analytics team, you are tasked to analyse historical data and develop predictive models for marketing purposes. Your manager has designed a pilot project focusing on clustering-based customer segmentation and profiling to discover consumer insights.

A number of analytics tasks are designed by the team to achieve the above objectives. You are expected to use SAS to perform clustering and profiling segments with the support of other tools like Excel for this assignment. You are required to relate the segments and profiles in conjunction with Roy Morgan value segments.

Assignment Structure

This assessment comprises one (1) case study with three (3) tasks. You must complete Task 1, Task 2, and Task 3.

Report Structure

Your report must address Task 1, Task 2, and Task 3 separately with clear section headings in a single PDF file. You may choose the format or style of the report you deem fit.

Dataset

Dataset files are available on LMS. The dataset is programmed to generate a random personalised subset based on your Student ID. Follow the instructions carefully to generate, save, and analyse your dataset before starting the tasks.

Case Study

Retail banks operate in highly competitive markets, where customer attrition (churn) can reduce profitability and long-term customer lifetime value. Banks therefore invest heavily in understanding why customers leave and how to proactively retain customers who are at risk of churn. A key challenge is that churn risk is rarely uniform across the customer base. Customers differ substantially in their demographics, financial behaviours, product relationships, and engagement patterns.

The bank has provided a dataset containing customer demographic information, account and relationship indicators, engagement/activity measures, and transaction behaviour. The dataset includes an outcome variable indicating whether a customer has churned.

As a customer analytics team member, you are required to use customer segmentation and clustering to identify meaningful customer groups and explain how these groups differ in churn risk and business value. You must also propose retention actions that are realistic and supported by evidence.

Requirements:

The project is seeking knowledge and insights relating to:
the demographics-based segments and their profiles;
the representative behavioural profiles for each segment;
how the produced segments can be mapped to broader consumer segment concepts in the Australian community.

A number of analytics tasks are designed by the team to achieve the above objectives. You are expected to use SAS Enterprise Miner to perform clustering and profile segments, with the support of other tools such as Excel where appropriate (e.g., for summarising results or preparing tables/figures).

You are required to relate your segments and profiles in conjunction with Roy Morgan Values Segments. Use the following link to further understand these value segments:
Roy Morgan Values Segments:

Note: The purpose of this mapping is interpretive. You are not expected to "prove" that your segments are identical to Roy Morgan segments. Instead, you should justify which Roy Morgan segment(s) your clusters most closely resemble, using evidence from your segment profiles.

Task 1: Demographic-based segmentation and churn profiling

Objective:
Develop a customer segmentation based on demographic and relationship characteristics and evaluate how churn differs across these segments.

You are required to:

develop a demographic-based segmentation using relevant demographic and relationship variables available in the dataset (for example, age, gender, education level, income category, marital status, dependent count, card category, and relationship indicators), and profile the resulting segments in a way that is meaningful for marketing and retention decision making.

Insight questions (answer all):

What variables did you select for demographic segmentation, and why were they appropriate for this task?

How many demographic segments did you identify, and what was your justification for this choice (e.g., interpretability, separation, business usefulness)?

How would you describe each segment in plain language (a short segment label plus key defining features)?

How does churn differ across demographic segments (e.g., churn rate by segment)? Identify the most at-risk segment(s).

What is one implication of these demographic segments for customer retention strategy (what would you do differently across segments)?

Based on your demographic segments, identify which segment(s) the bank should monitor most closely for churn risk and briefly explain why.

Recommend one segment-specific action (e.g., communication approach or product/service improvement) that could reduce churn for a priority segment.

Evidence requirement:
Include clear tables/plots showing segment profiles and churn outcomes by segment.

Task 2: Behavioural-based segmentation and churn profiling

Objective:
Develop a behavioural-based segmentation based on behavioural and engagement patterns and evaluate how churn differs across these segments.

You are required to:

develop a behavioural-based segmentation using relevant behavioural variables available in the dataset (for example, transaction amount, transaction count, utilisation ratio, revolving balance, inactivity indicators, contact frequency, and change measures),
you must map your segment(s) to Roy Morgan Values Segments and justify the mapping using evidence from your segment profiles.
interpret what these behavioural segments suggest about customer engagement and retention risk.

Insight questions (answer all):

What behavioural variables did you select for segmentation, and what do they represent in a customer engagement context?

How many behavioural segments did you identify, and what was your justification for this choice?

Provide plain-language profiles for each behavioural segment (include segment labels and defining characteristics).

How does churn differ across behavioural segments? Identify the highest-risk behavioural segment(s) and support your answer with evidence.

Identify at least one behavioural pattern that appears strongly associated with churn, and explain why it may indicate disengagement or attrition risk.

Identify which behavioural segment appears most at risk of churn and explain what behaviour signals disengagement.

Recommend one segment-specific retention action that is appropriate for this behavioural profile (and explain why it fits).

Evidence requirement:
Include clear tables/plots showing behavioural segment profiles and churn outcomes by segment.

Task 3: Combining segmentations and prioritising retention actions

Objective:
Combine demographic and behavioural segmentation insights to identify priority groups for retention strategy and propose actionable recommendations.

You are required to combine the segmentation results from Task 1 and Task 2 to generate deeper insight into churn risk and targeting. You must map your priority combined segment(s) to Roy Morgan Values Segments and justify the mapping using evidence from your segment profiles.

Insight questions (answer all):

Are there meaningful associations between demographic segments and behavioural segments? Explain what you observe (for example, which combinations occur most frequently and which are rare).

Which combined segment(s) represent the highest priority for retention, and why? Your justification must consider both churn risk and business relevance.

Under a targeting constraint (for example, marketing can only contact a limited number of customers), how would you prioritise who to target first? Explain your targeting strategy logic clearly.

Propose two retention actions for high-risk customers (at least one action should be segment-specific). For each action, specify:
who the action targets (which segment or combined segment),
why this action is suitable for that group, and
what KPI(s) you would track to evaluate success.

Propose one win-back action for customers who have already churned. Explain who you would prioritise and why.

Hint

If you choose to examine combined segments using Excel, you may use a pivot table that cross-tabulates demographic segment × behavioural segment × churn outcome. If you export segment membership from SAS, ensure each record can be matched correctly (e.g., customer identifier or row index).

Note: Assessment 2 only needed