Skip to content

Prajaktahz/Game-Analytics

Repository files navigation

🎮 Game Product Analytics: Player Engagement, Retention & Monetisation

Project Overview

This project analyses player behaviour in a turn-based multiplayer mobile game to identify opportunities for improving player retention, engagement, progression, and monetisation.

Using player activity, match history, progression, advertising, and purchase data, the analysis explores behavioural patterns across different engagement levels and translates them into actionable product recommendations.

Note: The original dataset is proprietary and is therefore not included in this repository. The notebooks and methodology have been retained to demonstrate the analytical approach.


Objectives

The analysis aims to answer the following business questions:

  • Where do players disengage from the game?
  • Which behaviours are associated with long-term engagement?
  • How does player progression influence monetisation?
  • What gameplay improvements could increase retention and revenue?

Data Preparation

Before conducting the analysis, several data quality checks were performed:

  • Validated data consistency across all datasets.
  • Corrected platform inconsistencies.
  • Investigated missing values.
  • Removed 371 records (0.015%) containing missing progression tier values, all of which belonged to 2v2 matches. As these represented only a negligible proportion of the dataset, their exclusion had no material impact on user-level analysis.

Key Insights

📉 1. Early Player Churn

A significant proportion of players disengage shortly after starting.

  • 37.6% of players were active for 5 days or fewer
  • Average player activity during the observation period was 18 days

Opportunity: Improve the early-game experience and onboarding to reduce initial churn.


📈 2. Engagement Drives Progression

Player engagement strongly correlates with progression.

  • Highly engaged players typically reach Tier 6
  • Lower-engagement players generally remain around Tier 3
  • Highly engaged players complete approximately 2.4× more matches per active day

Opportunity: Encourage repeat play through progression rewards and gameplay incentives.


🎯 3. Competitive Multiplayer Retains Players

Gameplay preferences differ substantially between engagement segments.

  • Highly engaged players spend ~90% of their gameplay in 1v1 competitive matches
  • Lower-engagement players spend more than 30% of their time playing against bots

Opportunity: Increase exposure to competitive multiplayer earlier in the player journey.


💰 4. Monetisation Improves with Engagement

Revenue increases consistently alongside engagement.

  • Highly engaged players generate 2.7× more advertising revenue
  • In-app purchase revenue also increases with player engagement and progression

Opportunity: Retaining players longer directly increases monetisation potential.


🚧 5. Mid-Game Progression Represents the Largest Opportunity

Progression analysis identified a clear bottleneck.

  • 21.5% of players are concentrated at Tier 6
  • Players progressing beyond this point demonstrate substantially higher engagement and spending
  • Player abandonment increases noticeably from Tier 8 onwards, indicating increased gameplay friction

Opportunity: Optimise the mid-to-late game progression experience to improve long-term retention.


Recommendations

Based on the analysis, several product improvements were proposed:

  • Improve onboarding and early-game progression to reduce initial churn.
  • Reduce progression friction around Tier 6.
  • Introduce additional rewards and progression support for advanced players.
  • Expand competitive multiplayer incentives to encourage long-term engagement.
  • Optimise advertising and in-app purchase strategies based on player engagement segments.

Repository Structure

├── notebooks/
│   ├── 01_Data_Exploration.ipynb
│   ├── 02_User_Engagement_Analysis.ipynb
│   └── 03_Progression_and_Monetisation.ipynb
├── README.md

Technologies Used

  • Python
  • Pandas
  • NumPy
  • Matplotlib
  • SQLite
  • Jupyter Notebook

Disclaimer

This repository showcases the analytical methodology and findings only.

The original dataset has not been included due to confidentiality requirements.

About

Analysing a turn-based tabletop sports game featuring competitive multiplayer gameplay, bot matches, advertising monetization, and in-app purchases. The objective of this analysis is to identify opportunities to improve player retention, engagement, and monetization through data-driven game design recommendations.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors