Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Calculating Customer Lifetime Value and Churn using Beta Geometric Negative Binomial and Gamma-Gamma Distribution in a NFT based setting

Sagarnil Das

TL;DR

This paper applies $BGNBD$ and $Gamma-Gamma$ distributions to NFT transaction data to estimate $CLV$. By modeling transaction frequency with $BGNBD$ and monetary value with $Gamma-Gamma$, the authors estimate $CLV$ from historical data and compare performance against traditional methods. Calibration-holdout validation and comparisons to $RFM$ and $Pareto/NBD$ show that $BGNBD$+$Gamma-Gamma$ better capture customer heterogeneity and improve predictive accuracy. The work provides practical guidance for marketing and retention in NFT ecosystems and highlights directions for real-time CLV estimation and incorporating engagement metrics.

Abstract

Customer Lifetime Value (CLV) is an important metric that measures the total value a customer will bring to a business over their lifetime. The Beta Geometric Negative Binomial Distribution (BGNBD) and Gamma Gamma Distribution are two models that can be used to calculate CLV, taking into account both the frequency and value of customer transactions. This article explains the BGNBD and Gamma Gamma Distribution models, and how they can be used to calculate CLV for NFT (Non-Fungible Token) transaction data in a blockchain setting. By estimating the parameters of these models using historical transaction data, businesses can gain insights into the lifetime value of their customers and make data-driven decisions about marketing and customer retention strategies.

Calculating Customer Lifetime Value and Churn using Beta Geometric Negative Binomial and Gamma-Gamma Distribution in a NFT based setting

TL;DR

This paper applies and distributions to NFT transaction data to estimate . By modeling transaction frequency with and monetary value with , the authors estimate from historical data and compare performance against traditional methods. Calibration-holdout validation and comparisons to and show that + better capture customer heterogeneity and improve predictive accuracy. The work provides practical guidance for marketing and retention in NFT ecosystems and highlights directions for real-time CLV estimation and incorporating engagement metrics.

Abstract

Customer Lifetime Value (CLV) is an important metric that measures the total value a customer will bring to a business over their lifetime. The Beta Geometric Negative Binomial Distribution (BGNBD) and Gamma Gamma Distribution are two models that can be used to calculate CLV, taking into account both the frequency and value of customer transactions. This article explains the BGNBD and Gamma Gamma Distribution models, and how they can be used to calculate CLV for NFT (Non-Fungible Token) transaction data in a blockchain setting. By estimating the parameters of these models using historical transaction data, businesses can gain insights into the lifetime value of their customers and make data-driven decisions about marketing and customer retention strategies.
Paper Structure (23 sections, 7 equations, 3 figures)

This paper contains 23 sections, 7 equations, 3 figures.

Table of Contents

  1. Introduction
  2. Background and Importance of Customer Lifetime Value (CLV)
  3. Overview of BGNBD and Gamma-Gamma Distribution Models
  4. BGNBD Model:
  5. Gamma-Gamma Distribution Model:
  6. Purpose of the Study and Research Questions
  7. Literature Review
  8. Advantages of BGNBD and Gamma-Gamma Distribution over Traditional Methods
  9. Methodology
  10. Description of the Dataset Used for the Study
  11. BGNBD Model and Its Parameters
  12. Gamma-Gamma Distribution Model and Its Parameters
  13. Results
  14. Model Performance Evaluation
  15. Comparison of BGNBD and Gamma-Gamma Distribution with Other CLV Estimation Methods
  16. ...and 8 more sections

Figures (3)

  • Figure 1: Frequency of Repeat Transactions (Model's prediction vs Actual)
  • Figure 2: Use of Calibration and Holdout Dataset for validating the BGNBD model
  • Figure 3: Actual Purchases vs Predicted Purchases