CASE STUDY

Leveraging Customer Relationship Data to Quantify Price Elasticity

ABSTRACT

Leaders at a US based financial services firm wanted to increase prices for one of their lending products. Given the aggressive competition in the market, they were reluctant to go for a blanket increase. Hence, they wanted to determine how they might leverage their customer relationship data to implement a differentiated pricing model. Identifying price insensitive segments allowed them to achieve their financial goals without incurring incremental attrition. 

 

Clearly Define Your Business Objectives

​A leading US-based financial services firm wanted to implement a new pricing strategy for one of its lending products using customer relationship data. This objective required analyzing the price elasticity of their customers, identifying variables showing correlation with price elasticity and developing a model using those variables to identify the customers for whom price can be increased without loss of revenue.

 

Acquire & Synthesize Relevant Data

​We began by collecting data from multiple sources, performing QA on the aggregated data and removing outliers to avoid skewed results. Afterward, we developed a metric to classify the customers as price sensitive or insensitive. Under that metric, we defined thresholds for different categories to keep a reasonable number of customers in each category. We subsequently flagged customers as ‘Price sensitive’ or ‘Price insensitive’ who showed consistent behavior across multiple years, as shown in Fig. 1. 

Fig. 1 – Categorization as Price Sensitive or Insensitive

We listed a number of independent variables, both quantitative and qualitative, that could affect price sensitivity and defined them at the beginning of the analysis period. We analyzed the behavior of each variable with respect to price elasticity and shortlisted the ones which showed the highest correlation.

 

Develop an Action Plan

The customers needed to be segmented as either ‘Insensitive’ (those for whom price can be increased without any loss of revenue) or ‘Sensitive’ (those for whom price increase can lead to loss of revenue); out of the shortlisted variables, we picked variables without correlation to each other. Using these as ‘nodes’, we created a Decision Tree (Fig. 2).

Fig. 2 – Customer Segmentation

Having identified the ‘Insensitive’ customers, we created sub-segments for them based on their volume and recommended that price be increased to their segment average.

Partner with You to See it Through

 

Partner with You to See it Through

Using this model, our client was able to identify the ‘Insensitive’ customers and implement a differential pricing model whereby price for a customer was changed depending upon their respective segments. The new pricing model helped our client drive revenue growth of over 20% without substantial increase in customer attrition rate.

Keywords 

Price Elasticity, New York Consulting, Management Consulting, Financial Services, Commercial Lending, Data Analytics, Quantitative Analysis, Analytics, Case Study

 

Back to ANALYTICS

OUR CASE STUDIES

Discover how our 4-step process has helped clients avoid confusion and succeed in the marketplace.

ABOUT

CAREERS

INSIGHTS

CONTACT

Copyright © 2019 FischerJordan. All Rights Reserved.