Project Type: B2C, mobile application (iOS, Android, Flutter), Computer Vision and Machine Learning technologies
Brief Description: Development of a unique technology for analyzing athletes' techniques (specifically golfers) using computer vision algorithms and a biomechanical system. The application provides personalized recommendations and enhances user experience, helping athletes correct their technique without the need for comparison with a standard.
Key Project Features
Goals and Objectives:
Development and implementation of AI algorithms for personalized recommendations
Conducting user research and gathering feedback
Developing partnership strategies with golf clubs and coaches
Improving the user interface and functionality of the application
Accelerating training progress and reducing the risk of injuries
Key Results (KPI):
Reducing the time for technique analysis by 40%
Reducing the number of injuries by 30%
Increasing training efficiency by 25%
Increasing the number of active users (MAU) by 50% in the first 6 months
My Role and Team
My Role: Senior Product Manager
Main Responsibilities:
Developing long-term product strategy and unique selling propositions
Implementing and testing AI algorithms to improve user experience
Conducting user research and analyzing metrics
Developing partnership strategies with golf clubs and coaches
Improving the user interface and functionality of the application
Developing and implementing unique technology for athlete technique analysis
Team Composition:
1 PM
4 Developers
2 Analysts
2 Testers
Timeline and Tools
Timeline: 2020 – 2024
Project Management Tools:
Using Agile and Scrum methodologies
Jira for task tracking
Confluence for documentation
Slack for communication
Figma for creating CJM and analyzing customer interactions
Achievements and Results
Successful Moments:
Successful implementation of the technology in sports schools and clubs
Positive feedback from coaches and athletes
Increasing the number of users by 50% in the first 6 months
Securing 20 million rubles in project funding
Project Metrics:
CAC (Customer Acquisition Cost): $50, due to increased marketing campaign volume
LTV (Lifetime Value): $300, clients started using paid features and renewing subscriptions
Revenue: $18,000, due to user growth and the introduction of paid features
ARPU (Average Revenue Per User): $30, due to the implementation of paid subscriptions and services
MAU/DAU (Monthly Active Users / Daily Active Users): 100 / 25, due to increased marketing efforts and improved user experience
User Growth Rate: 20%, due to active marketing campaigns and referrals
Issues and Solutions
Computer Vision (CV) Issues
Problem: High-quality input video data requirements for correct athlete technique analysis.
Solution:
Developing and implementing algorithms capable of processing low-quality video and highlighting key biomechanical markers.
Conducting numerous tests and improvements to enhance algorithm accuracy and resilience to external conditions.
Using machine learning methods to adapt algorithms to various shooting conditions, such as lighting and angles.
Problem: Limited computational power of mobile devices to work with heavy computer vision algorithms.
Solution:
Optimizing algorithms for efficient operation on mobile devices using Flutter technology for cross-platform development.
Implementing distributed computing, where part of the analysis is performed on the server instead of the mobile device.
Regular updates and code improvements to enhance performance and reduce power consumption.
Development Issues
Problem: Difficulty in integrating multiple functions and algorithms into a single application with an intuitive interface.
Solution:
Conducting detailed user research and creating CJM in Figma to analyze customer interactions.
Working closely with designers to develop a user-friendly and understandable interface.
Continuously collecting user feedback and iteratively improving the interface and functionality based on this information.
Problem: Supporting stable application operation on various platforms (iOS, Android) and devices.
Solution:
Using Flutter for cross-platform development, ensuring a single codebase for all platforms.
Thorough testing on various devices and platforms to identify and fix bugs.
Implementing automated testing for quick problem detection and reducing code errors.
Communication Issues with Clubs and Coaches
Problem: Clubs and coaches' distrust of new technologies and their effectiveness.
Solution:
Organizing demonstrations and presentations showing real examples of technique improvement using the application.
Conducting pilot projects in partner golf clubs to obtain real feedback and recommendations.
Constant interaction with coaches and clubs, discussing their needs, and integrating their suggestions into the product.
Problem: Challenges in establishing and maintaining long-term partnerships with clubs and coaches.
Solution:
Developing a partnership strategy that includes favorable cooperation conditions and bonus programs for clubs and coaches.
Regular meetings and seminars to share experiences and discuss opportunities for improving the application.
Continuous support and training for coaches on using the application for athlete technique analysis.