Overview
Data Pointes Lab is a passion project that makes pointe shoe data clear, accessible, and useful. Built and maintained by a dancer, engineer, and data scientist — the platform combines personal experience in ballet with technical expertise to create a data tool that helps dancers, parents, teachers, fitters, and researchers make informed choices
Platform Highlights
330+ shoe models with structured specifications
Transparent filtering system – every algorithm is visible and explainable
Similarity Finder – shoes compared across 8+ attributes (box shape, shank strength, vamp, platform width, heel height, material, etc.)
Barre Graphs – dynamic charts for brand trends, pricing insights, and attribute distributions
Relevé Relatives – network graph showing how models cluster and connect
Size Variations – brand-specific street-to-pointe size converter
Open access & open source – database is free to explore and downloadable for research
How It Works
Data Pointes Lab is a comprehensive pointe shoe data platform that combines transparency, accuracy, and usability. At its core are four modules: Corps de Data (database of 330+ models with advanced filtering), Size Variations (cross-brand sizing tool with padding adjustments), Relevé Relatives (interactive network of shoe relationships), and Barre Graphs (visualizations of trends and characteristics).
Behind the scenes, the system uses a graph-based similarity algorithm that compares shoes across 8+ attributes — including box shape, platform width, shank strength, vamp length, and foot shape. Each factor has a clear mathematical definition (from simple categorical matches to Jaccard similarity and custom similarity matrices), and weights can be adjusted in real time. A live equation display shows exactly how similarity scores are calculated, ensuring full transparency.
The platform’s data architecture is built from official brand specifications, cross-checked retailer data, and community feedback. Models are continually updated, with discontinued items tracked alongside current availability and pricing. Interactive visualizations use force-directed layouts and responsive design for intuitive exploration on desktop and mobile.
All data is open and downloadable and in addition to the platform, it can be found on the GitHub Repository
To support this platform, Heymann Apps also developed a data collection and validation tool. This agentic system leverages APIs (e.g., Brave search), modern LLMs (like Anthropic Claude), and custom pipelines to assist with gathering and verifying pointe shoe specifications. Combined with manual research, this ensures the database stays accurate, current, and scalable. See it in action below.