Applied AI in sports analytics
Challenge
Atavus Sports wanted to scale and automate its analytics for tackling performance using AI. While it had deep domain expertise and valuable video and field data, its processes were manual and difficult to scale, limiting its ability to generate insights efficiently.
Our thinking
We recommended a phased approach combining data analysis, statistical modelling, and rapid experimentation with AI. By first strengthening the data foundations and key metrics, then using focused “hack” sprints to test machine learning and computer vision approaches, Atavus could quickly identify what would deliver real value and avoid wasted effort.
Execution
Matchbox led discovery across data, processes, and infrastructure, working with partners to refine key performance metrics and build a solid analytical foundation. We then collaborated with Microsoft and Management Foresight in a rapid prototyping sprint, using cloud-based AI tools to test computer vision and machine learning approaches for automating analysis.
Outcome
Atavus gained a clearer data strategy, improved analytical models, and validated approaches for scaling its platform with AI. The project also demonstrated how rapid experimentation and strong data foundations can accelerate innovation while reducing risk.