FLO aims to create a strategic roadmap for its sales and marketing activities.
To support medium- and long-term planning, it is essential to estimate the potential future value (Customer Lifetime Value) that existing customers will generate.
This enables:
- More accurate budgeting and forecasting
- Data-driven marketing strategies
- Improved customer relationship management
The dataset consists of customer shopping behavior data from 2020–2021, covering OmniChannel activity (both online and offline).
| Variable | Description |
|---|---|
| master_id | Unique customer ID |
| order_channel | The channel/platform used for shopping (Android, iOS, Desktop, Mobile, Offline) |
| last_order_channel | The channel of the most recent purchase |
| first_order_date | The date of the first purchase |
| last_order_date | The date of the most recent purchase |
| last_order_date_online | The date of the most recent online purchase |
| last_order_date_offline | The date of the most recent offline purchase |
| order_num_total_ever_online | Total number of online purchases |
| order_num_total_ever_offline | Total number of offline purchases |
| customer_value_total_ever_offline | Total amount spent on offline purchases |
| customer_value_total_ever_online | Total amount spent on online purchases |
| interested_in_categories_12 | Categories shopped in during the last 12 months |
- Estimate Customer Lifetime Value (CLTV) using probabilistic models
- Predict future purchasing behavior
- Identify high-value customers
- Support targeted marketing strategies
- Data preprocessing and feature engineering
- Creation of CLTV dataset structure
- Model implementation:
- BG/NBD (Beta-Geometric / Negative Binomial Distribution) → purchase frequency prediction
- Gamma-Gamma Model → monetary value prediction
- CLTV calculation (6-month projection)
- Customer segmentation based on CLTV
- Accurate estimation of customer lifetime value
- Identification of high-potential customer segments
- Improved marketing efficiency and ROI
- Better long-term strategic planning