Skip to content

Lloyd's Banking Group Data Science

Image title

Welcome to the Data Science Job Simulation

Your Role

  • You are a data science graduate at Lloyds Banking Group.
  • You are part of the Data Science & Analytics team, where each member contributes uniquely to projects.
  • Your team has been assigned the task of enhancing customer retention through analytics, focusing on predicting customer churn.
  • As a recent graduate eager to apply your data science skills, you see this project as an excellent opportunity to make a meaningful impact.

Your Goal

  • Your primary task is to collaborate with the team to gather and analyse data relevant to customer churn.
  • You need to perform exploratory data analysis to uncover patterns and insights, clean and preprocess the data, and build a predictive model for customer churn.
  • Additionally, you are expected to suggest ways to evaluate and measure the model's performance.

Project briefing

Project brief

Welcome to the Data Science & Analytics team at Lloyds Banking Group. As a new data science graduate, you have been entrusted with a critical project that could significantly impact our customer retention strategies. Li, our senior data scientist, has specifically chosen you to take over this project due to your strong analytical skills and enthusiasm for solving real-world business problems. This is an exciting opportunity for you to apply your knowledge and make a real impact within our team.

Context

The project you are about to embark on is the "Customer Retention Enhancement through Predictive Analytics" initiative. This project arose from an urgent need to address declining retention rates within certain segments of our customer base. Over the past few months, we've noticed a worrying trend of increased customer churn, particularly among young professionals and small business owners. This poses a substantial threat to our market position and long-term profitability.

Our fictional client, SmartBank, a subsidiary of Lloyds, has reported that a substantial portion of their customer base is at risk of moving to competitors offering more personalised banking solutions. SmartBank has tasked our team with developing a predictive model to identify at-risk customers and propose targeted interventions to retain them.

Key concepts

Before you begin, it's essential to understand a few key concepts:

  1. Customer churn: The process by which customers stop doing business with a company. Identifying and preventing churn is crucial for maintaining a stable customer base.
  2. Predictive analytics: Techniques that use historical data to forecast future possibilities. In this project, you'll use predictive analytics to predict which customers are likely to churn.
  3. Exploratory data analysis (EDA): A method of analysing data sets to summarise their primary characteristics, often using visual strategies. EDA is crucial for understanding the data before building predictive models.
  4. Machine learning models: Algorithms that let computers learn from and make predictions or decisions based on data. You'll be building a classification model to predict customer churn.

Project requirements

Your task involves two main phases.

Phase 1: Data gathering and exploratory analysis

  • Objective: Collect relevant data, perform exploratory data analysis (EDA), and prepare the data for modelling.
  • Steps:
    • Identify and gather data from provided sources relevant to predicting customer churn.
    • Conduct EDA to understand the data, identify key features, and uncover patterns and insights.
    • Clean and preprocess the data to ensure it is ready for model building.
  • Deliverable: Submit a report detailing your exploratory data analysis and key findings, including the cleaned and preprocessed data set.

Phase 2: Building a machine learning model

  • Objective: Develop a predictive model to identify customers at risk of churning and propose ways to measure the model’s performance.
  • Steps:
    • Choose a suitable machine learning algorithm for the classification task.
    • Build and train the model to predict customer churn.
    • Suggest ways to evaluate and measure the model’s performance, ensuring its reliability and accuracy.
  • Deliverable: Submit a report, including the trained machine learning model and your proposed methods for evaluating and measuring the model’s performance.

The importance of this project

Addressing customer churn is critical for maintaining SmartBank's competitive edge and ensuring long-term sustainability. The board of directors is strongly encouraging the team to deliver a solution promptly, as timely delivery will help secure our market position and capitalize on opportunities. The team is feeling the heat, but they are also highly motivated and determined to turn this challenge into an opportunity. Your role is pivotal in this effort, and your contributions will be closely monitored by senior management.

Embrace this chance to showcase your skills and make a meaningful impact. It’s time to dive in and get to work!

Tasks & Notebooks

The task descriptions and jupyter notebooks for this project

Any questions or comments about the above post can be addressed on the mldsai-info channel or to me directly shtrauss2, on shtrausslearning or shtrausslearning