MY ROLE
MY ROLE
Embedded Files
Business Analyst, Computer Science Engineer
DATE
DATE
Mar 2019 - May 2019
TEAM MEMBERS
TEAM MEMBERS
Shreya Choudhury, Shariq Farhan
TOOLS
TOOLS
R Studio, SQL Server Management Studio
BUSINESS NEED
BUSINESS NEED
After the 2016 merger of Dell and EMC, Dell faced an expansion in its product portfolio, resulting in an increased customer base and diverse buying patterns. The availability of abundant data provided an opportunity to analyze and understand customer behavior.
THE SOLUTION
THE SOLUTION
To gain deeper insights into customer interactions and segment them based on transaction patterns, Dell initiated the Customer Segmentation project. By utilizing the clustering capabilities of R and employing a wide range of numerical and categorical features, we successfully identified and categorized account personas based on specific business parameters.
The project began with a successful North American pilot, which was later scaled to all regions, for various use cases.
PROCESS
PROCESS
DISCOVER
DISCOVER
Requirements Gathering
Requirements Gathering
As part of the requirements-gathering process, we engaged with Stakeholders and the internal pricing analytics team to learn how current product launch strategies were devised and implemented.
Requirements gathering
Insights
Insights
From the discussions, the following key insights emerged.
DEFINE
DEFINE
The Customer Segmentation project directly addresses all the above-mentioned problems. It provides effective analysis and solutions by utilizing comprehensive data analysis techniques to segment customers based on their transaction patterns. This enables targeted marketing and allows for better resource allocation, optimizing customer engagement, and maximizing business opportunities.
For a 360-degree view of customer characteristics, both categorical and numerical parameters were considered that included but were not restricted to:
IMPLEMENT
IMPLEMENT
The broad phases of the implementation process involved the following steps.
RESULTS
RESULTS
The clustering and profiling exercise resulted in ten distinct clusters, each with a unique persona.
The clusters yielded intriguing and valuable insights.
The findings were then shared with respective teams for their use.
Example of two clusters with similar hardware preferences but very different personas are shown below.
The Customer Segmentation model is currently being utilized by Dell as a live project. After the success of the pilot project, the model was extended to use cases in the Marketing and Technical Support teams.
Page updated
Report abuse