Who is the Client

A US-based skincare products manufacturer, with $1.5B annual sales and over 2 million customers in the United States, Canada, and Australia.

The Challenge

The client distributes its products through a multi-level marketing model. They hire independent sales consultants that connect with potential preferred customers (PCs) via social media or in-person meetings. The PCs eventually buy and resell the products to the end customers. They play a vital role in pushing sales and they place a certain amount of order every 60 days, which forms a major part of the client’s sales.

If, for any reason, these PCs start becoming inactive, it can result in a huge business loss for the client. The inactive PCs also impact the consultants’ commissions and jobs, causing more harm to the client’s entire sales network. The client wanted to develop a model that can classify PCs into segments based on their engagement levels. It wanted to target these segments with personalized emails to retain the churning PCs by offering them exciting deals.

The Solution

GSPANN’s advanced analytics team created the model to predict customer churn utilizing its in-house ML solution called ‘Zeolite’ that is based on Keras (Python deep learning library). We leveraged data residing in the existing SQL servers and combined it with Azure cloud’s processing and deployment capabilities to provide a feasible solution.

After running successful POCs on various deployment methods, the model was finally deployed on Azure ML compute nodes and an automated batch-mode scoring pipeline was built. Further tuning and re-training was done in the second version of the model with additional features to achieve more accurate results.

Various data types with possible relations with the churning PCs were collected. These include net sales, order counts, tenure, email-click stream data, sponsor attributes, last purchase details, etc. The prediction made through the Deep Learning Keras Sequential Model showed around 73% of training and test accuracy, and 76% of Recall (the ability to find all the positive samples).

The model was deployed on Azure ML platform and a scoring pipeline was created wherein new data files can be scored in bulk with the help of an API POST call. The model scores the probability of each churning PC on the scale of 0 – 1. Based on the probability, the client segments the PCs into 10 segments, picks 3-5 segments with the highest probability, and targets them with personalized offers. The model is re-trained every 2-3 months. The solution also analyzes the results of the discount campaign that is run on the target customers.

Business Impact

  • To analyze the success of the model, 60K PCs were picked based on their probability score and were divided into two groups: Targeted Group (50K PCs who were sent the promotional offers) and Control Group (10K PCs who were not sent any promotional offers).
  • The email campaign performed well on the targeted group. The results showed 23.7% Open Rate; 3.2% CTR (Click Through Rate); and 13.4% CTOR (Click to Open Rate).
  • Following are the targeted group performance metrics compared to the control group during the 2-week campaign.
    • The targeted group sales were 82% higher compared to the control group, average order value (AOV) was increased by 9%, purchase rate was increased by 60% compared to the control group, and average revenue per customer was 80% higher compared to the control group.
    • After the campaign was over, we observed the following trends for the next 45 days: The targeted group sales were 9% higher compared to the control group and purchase rate was increased by 11% compared to the control group.

Technologies Used

Zeolite. GSPANN’s suite of ML solutions
Python. An interpreted, high-level and general-purpose programming language
Environment. Windows, Azure Compute Nodes

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