Retail User Segmentation using RFM model

Worked with retail chain to segment their customers based on RFM (Recency, Frequency, Monetary) model inorder to target its loyal users. It's particularly useful in understanding customer value, improving targeting strategies, and maximizing returns on marketing efforts.

1. Data Collection & Preparation:

Extraction of transaction data for one of outlets of grocery chain along with inputs necessary to calculate recency, frequency and monetary value for a given time period.

Convert data into a table where each row represents a customer, with columns for recency, frequency, and monetary metrics.

2. Evaluation:

Assign scores(say 1-5) to each customer based on their RFM metrics.

The highest score (5) represents the top 20% of customers for each metric, and the lowest score (1) represents the bottom 20%.

Combine RFM scores to form an overall customer score (e.g., a customer with scores of R=4, F=5, M=5 is highly valuable).

3. Segmentation:

Group customers based on their RFM scores.

Typical segments include:

1. Market Campaign & A/B testing:

Based on RFM models, champions and loyal customers of the platform were picked to provide rewards and referral options exclusively through e-mail marketing to encourage engagement and conversions.

This was followed by A/B testing to test the success of the campaign

2. Interactive Dashboard:

A dashboard displaying the distribution of RFM scores, segment performance, and customer behavior patterns.

Filters for sales teams or marketing managers to drill down into specific segments or time periods.

Some key metrics include Customer Lifetime Value (CLV), average order value (AOV), customer retention rate per segment, etc.