Case Study - Attitudinal Segmentation using Transaction Data

Situation

A large Australian retail bank aimed to:

Optimise use of their marketing resources
Increase cross-sales of products to their existing customers.

External research had shown that organisations can realise significant gains in sales and marketing efficiency by identifying customers who viewed them as their primary financial provider, and are more likely to respond to proactive targeted offers.

Recent campaigns had shown that clients with at least one major product had a many times higher likelihood of responding to offers than those with no major products.

Vision

The bank aimed to find how related customers felt to the bank and how this affected current and future profitability of the customer. They would then be able to define marketing strategies tuned to the customer's future needs, likelihood of responding and future value.

Specific Requirement

The aim was to create a ‘Strength of Relationship’ measure for all customers indicating the attitude of the customer towards the bank and their likelihood of purchasing more products. The data available to perform this was the customer, product and transaction data held by the bank. There were no previous indicative attitudinal measures available for any customers that could be used as a basis.

We Provided

Research into segmentation techniques and tools suited for the task.
Running of brainstorming sessions with business and marketing experts to determine possible indicative attributes and criteria to determine success.
Collection of all available data attributes that were defined as indicative.
Conversion of data fields into meaningful data attributes.
Experimentation with different segmentation structures in order to find the most informative and useable segmentation that satisfies the defined criteria of success.
A set of rules separating customers into groups in order of ‘strength of relationship’ suitable for targeting by different marketing strategies.
Profiles of the different segments by many different attributes in order to create a vision of the customers in each segment.
Full documentation of the whole project including modelling methodology, diagnostics and results.
End-to-end code for monthly runs of the scoring process, including comments and built in quality checks.
Documented observations and learnings from the project.

Result

The ‘Strength of Relationship’ measure is used on an ongoing basis to perform targeted campaigns with increased response rates and reduced contacts, resulting in increased profit.

The segmentation separates customers into 10 groups in increasing “Strength of Relationship” measures. The customers in the top segment have a history of responding to marketing campaigns 5 times more than the average customer, compared to 1/5^th of the time for the bottom segment. Focus groups showed that customers in the same segment had generally the same view towards the bank in terms of their current satisfaction and future banking needs, and views between segments were very different.

There is now an understanding of the factors that indicate customer satisfaction and strategies have been developed to strengthen the relationship of a customer with the bank.