Who is the Client

A US-based retailer with more than 700 departmental stores and 150 specialty stores, client is a major player in their industry and one of the oldest shopping destinations for their customers.

The Challenge

The client runs a huge e-commerce fashion portal that also has beauty products. They were sending email reminders manually to every customer, in an interval of 60 days, for reordering beauty products based on the date of purchase.

However, due to the lack of personalization and incorrect delivery timings of the reminder emails, customers treated them as spams. As a result, their click-through rate (CTR) went low, rendering the whole e-mail marketing campaign for reorder sales ineffective.

The client wanted to increase sales of their beauty products by sending personalized and focused email reminders. The challenge was to estimate the correct reorder interval to send timely, personalized e-mail notifications to each customer for all beauty products.

The Solution

GSPANN’s Advanced Analytics Team created the ML algorithm utilizing its ‘Zeolite’ solution to predict when a customer is more likely to rebuy the same beauty product. For this purpose, we trained the ML model with historical data of customers. Before preparing the model, we performed multiple subsidiary tasks, including data extraction from various data sources, data exploration and research, data cleaning and wrangling, model building, and deploying the model in the client production system.

We compared model performance with the previous scheme by running A/B tests for six months. After ensuring performance superiority and consistency, we deployed it in production. Our Zeolite solution extracts three dimensions of data – customer data, product data, and transaction data stored in the client’s databases. We used Python scripts to clean data and create features required for model training using the “Gradient Boosted Machines” algorithm and predictions for everyday transactions.

The solution utilizes scheduled Python scripts for extracting data on a daily basis with the Control-M scheduler. Next, we transferred data to the saved model for predictions. Finally, we stored forecasts from the model in a new table of the database.

Business Impact

  • The ML model learned from the data of 47+ million transactions that happened over two years. As a result of smart reminder emails, the client observed a 4% increase in email open rate and a 14% increase in CTR when compared to the previous year.
  • The Average Order Value increased by 16%, and the email un-subscription rate dropped by 23%.
  • The client estimated a sales increase in the beauty segment by $400K in the year implemented.

Technologies Used

Zeolite. An AIOps framework that utilizes artificial intelligence and machine learning to detect potential failures automatically
Jupyter. An open-source web application that allows you to create and share documents containing live code, equations, and visualizations
Python. An interpreted, high-level, and general-purpose programming language that enables programmers to write clear and logical code
Control M. It makes it easy to build, define, schedule, manage, and monitor production workflows, ensuring visibility, reliability, and improving SLAs

Related Capabilities

Utilize Actionable Insights from Multiple Data Hubs to Gain More Customers and Boost Sales

Unlock the power of data insights buried deep within your diverse systems across the organization. We empower businesses to effectively collect, beautifully visualize, critically analyze, and intelligently interpret data to support organizational goals. Our team ensures good returns on the big data technology investment with effective use of the latest data and analytics tools.

Do you have a similar project in mind?

Enter your email address to start the conversation