QBUS6600 Project 1 Outline: L'Oréal Dermatological Beauty - Predicting Customer Transaction Value

Background

L'Oréal Dermatological Beauty (LDB) is a division of L'Oréal dedicated to providing dermatological solutions for various skin concerns. With brands like La Roche Posay, CeraVe, and SkinCeuticals, LDB offers a wide array of products designed to address specific skin issues such as acne, eczema, and more. While LDB products are primarily sold through pharmacies, the division is focusing on expanding its online presence and sales through its brand-dedicated online stores, such as the one for La Roche-Posay.

Selling through the online store presents numerous advantages, including the ability to offer a more personalised experience to customers, or as LDB calls them, "patsumers1 ". This personalised approach aims to match products to individual skin concerns and preferences, enhancing the customer journey and fostering brand loyalty. Moreover, the online platform. allows for constant communication with patsumers, keeping them informed about new product launches and promotions. For example, emails are sent to everyone who has signed up to the LDB mailing list, with links to suggested products.

LDB is committed to enhancing its personalisation strategy by leveraging customer data to better predict the future transaction value of its patsumers. By analysing purchasing history and browsing habits, LDB can gain valuable insights into customer behaviour and preferences. This approach allows LDB to anticipate patsumers' needs more accurately, enabling more effective targeting and personalisation of marketing efforts. The potential benefits include increased customer satisfaction, higher retention rates, and improved sales performance.

Problem Description

Use the available data (see ‘Data Description’ below) to develop a model for predicting the transaction value of a patsumer over the next six months. You can frame. this task as a regression problem for predicting the monetary value of future transactions for each patsumer. The project presents a unique opportunity to apply your data analytics skills to a real-world business challenge and contribute to the ongoing success of L'Oréal Dermatological Beauty.

In this project, you should:

• Use Exploratory Data Analysis (EDA) to identify the top behaviours and attributes that are likely to predict the transaction value of a patsumer.

You should aim to find or reveal all relevant properties, characteristics, patterns, and statistics hidden in the datasets.

• Develop a regression model for predicting the transaction value of a patsumer over the next six months.

Implement any statistical or machine learning approaches that you feel are appropriate. Ensure that you justify the selection of your model and interpret the model in terms of the key attributes for predicting the future transaction value. Use the RMSE to evaluate the performance of your final model.

• Based on your analysis, outline a strategy to help LDB prioritise customers, enabling better targeting and personalisation of marketing efforts and services, and to improve sales performance.

You should design a specific potential project for the LDB team to execute, to take advantage of the key behaviours and attributes that you have identified, and the models you have built and validated. The project could include marketing programs, product enhancements and/or other interventions or changes. The project should be backed by high-level revenue and cost estimates, accompanied by assumptions and/or supporting data.

Data Description

You have been provided with a tabular dataset in CSV format (~6.4K rows) encompassing transactional data (sales history) from November 2023 to May 2024. This dataset includes information on individual transactions, product details, and descriptive features of the patsumers, such as their previous purchases, and other relevant attributes. LDB has also included a data dictionary that summarizes each column in the dataset.

LDB has made an effort to ensure the data is relatively clean, however, we encourage you to perform. checks and conduct the necessary data processing and feature engineering. You are also welcome to explore external datasets to enrich your analysis and feature engineering.

Useful Tips

• Data Dictionary: Review the provided data dictionary to understand each variable.

• Train-Test Split: Implement a train-test split to validate your model's performance and prevent overfitting.

• Feature Engineering: Perform. feature engineering to enhance model performance. Creating and transforming features can uncover hidden patterns.

• Experiment with Models: Test various regression models to find the most suitable one. This experimentation is key to achieving high model performance.

• Real-Life Scenario Application: The dataset includes only customers who made a transaction in the past 6 months (less than 20% of all customers). Discuss applying this model in real-life scenarios, such as a two-stage model (first stage: binary classification to predict transaction likelihood; second stage: regression to predict transaction value). This approach enhances practical applicability.