Cloud Pak for Data

 View Only

Cloud Pak for Data accelerators in wealth management | Customer Segmentation

By Doug Moran posted Tue July 09, 2019 07:44 PM


The following videos show how you can use the Customer Segmentation accelerator in IBM Cloud Pak for Data to go beyond the traditional demographic markers, such as age, income, and gender to identify meaningful customer segments. 

Jenny, a data scientist, is tasked with identifying more dynamic customer segments for her company so that they can better serve their client base. She decides to use the Customer Segmentation accelerator to get started quickly.

Customer segmentation prediction - Overview video  

This video provides an overview of the accelerator and describes its relevance to the wealth management industry. Curating meaningful customer segments is critical for success in the wealth management industry, because it enables you to identify customers you might otherwise miss and create truly tailored marketing campaigns.  Go beyond traditional demographic markers by uncovering common behavioral patterns.

Customer segmentation prediction – Data catalog video 

Before creating a Customer Segmentation model, Jenny, the data scientist, needs to have access to the right data. This video shows the business glossary that is included in the Customer Segmentation accelerator, and explains how the terms in the glossary are mapped to your data when a data steward runs automated discovery on the data sources that are connected to Cloud Pak for Data. The video demonstrates how to use the relationship graph to explore the relationships between the terms. The curated information architecture makes it easier for Jenny to identify the right data to use in her analysis. 

Customer segmentation prediction – Data prep & ML model training video 

This video shows how Jenny, the data scientist, can use a sample Jupyter notebook to prepare her data and build and train a machine learning model to identify customer segments. The notebook includes information about the script that she can run to cleanse and prepare her data and standardize her variables. The notebook also helps her use K-means clustering to partition the data into seven clusters and then visualize the data in the clusters.

Customer segmentation prediction – ML model scoring video 

Now that Jenny, the data scientist, has trained her machine learning model, she's ready to test and score the customer segmentation model. This video shows how she can use a sample Jupyter Notebook that walks her through the process of testing and scoring the customer segmentation model and shows her how to deploy the model as a web service. 

Customer segmentation prediction – User API interface video  

This video shows how Jenny, the data scientist, can use the deployed web service and the sample dashboard in the accelerator to leverage the insight generated by the customer segmentation model. The dashboard enables her to see the different customer segments and the clients who are aligned with each segment. The video also shows how to deploy the dashboard so that other employees at the company can access the web application to identify new and dynamic segments of customers to develop more personalized outreach and results.