The goal of this projectis to apply all data analytics concepts that I have learned so far to real-world data both to practice and better understand how these concepts are useful.
This project is about the Customer Lifetime Value (CLV) of the customers of the webshop Yermana.
CLV is used to measure the performance of returning customers. CLV is the expected profit that is generated from a customer over the course of a business relationship. The longer a customer purchases from a company, the greater the customer lifetime value becomes.
Why? That information can be used to segment customers into profitable and less profitable customers. Based on this segmentation, the allocation of marketing resources or investment in customer retention can be altered. This will help a business to grow profitably in the long-term.
CLV = Lifetime * Purchase Frequency * AOV * Profit Margin
-
Lifetime is the expected duration of an engaged customer (activity rate, churn, cohort consideration, repurchase rate). There are probabily one-time customers and regular customers.
-
Purchase frequency shows how often a customer purchases in a lifetime. The time between purchases can provide a good approximation as to the purchase frequency. Purchase Frequency ≈ Purchase Frequency Time Interval / Time Between Purchases Purchase Frequency = Total Number of Orders / Total Number of Unique Customers
-
Average Order Value (AOV) is the average of each basket and is calculated as follows: AOV = Unit / Order * Price / Unit
-
Profit Margin is the ratio of revenues minus costs to revenue. It is important to take into account the attribution and allocation of costs and revenues. For example, Google Analytics attributes based on a last-click basis.
-
Churn is a measure loyalty of a customer
-
Recency, Frequency and Monetary (RFM) segmentation has been used as a means to estimate CLV. This might be interesting to look into.
Check if Customer Acquisition Costs (CAC) is lower than Customer Lifetime value
How can we estimate Customer Lifetime Value (CLV) and what are the driving factors of CLV?
The dataset that is used contains data from 2016 to 2021.
- Project planning
- Question formulation
- Data cleaning
- Data storing
- Data exploration
- Hypothesis formulation
- Hypothesis testing
- Formulation results
- Preperation presentation
I've created a planning with Trello
Yermana Research - Harvard Business Review Research - CLV Metric Research - Questions about CLV Research - CLV in E-Commerce Research - Increasing CLV Research - CAC < CLV