onefit.ai: Helping people buy the right shoes online without compromising privacy

This case study explores how onefit.ai, an AI-driven startup, is tackling one of the most persistent challenges in online retail: helping people buy shoes that actually fit when they arrive on the doorstep. By using computer vision and machine learning techniques to estimate foot measurements from a single photograph, onefit.ai aims to reduce unnecessary returns, improve customer confidence, and lower the environmental impact associated with online footwear shopping. Taking part in The Alan Turing Institute’s Turing Way Practitioners Hub has helped onefit.ai achieve its commitment of embedding responsible AI practices from the outset – particularly around data minimisation, privacy and user trust. This case study is published under The Turing Way Practitioners Hub 2025-26 Cohort - case study series. The Practitioners Hub is The Turing Way project that works with experts from partnering organisations to promote data science best practices. Key takeaways Poor fit remains a major barrier to effective online footwear retail, driving up product returns, waste and lost sales. AI-based technology can improve sizing accuracy and sales – but only if retailers are persuaded to adopt and users trust how their private data is being handled. Responsible AI practices such as data minimisation and transparency are a key consideration in consumer-facing applications. Representatives of retail companies may be sceptical of AI or lack knowledge of the technology – and this needs to be treated with empathy and not misinterpreted as hostility. Building an evidence base around the effectiveness of an AI system, as well as aspects such as its environmental impact, is important but not always straightforward. The Turing Way Practitioners Hub provided reassurance, peer learning and practical tools for navigating shared AI challenges.