Case Study: Hindustan Unilever Limited achieves scalable ML-driven consumer insights and demand forecasting with HackerEarth

How the world’s most innovative FMCG company crowdsourced Machine Learning models to understand consumer preference

Hindustan Unilever Limited (HUL), through its humarashop.com e‑commerce initiative, sought to understand changing consumer preferences in small neighborhood retail stores by capturing sales data from point‑of‑sale systems and applying Machine Learning. HUL’s challenge was finding a sophisticated open‑innovation platform and access to a large, active ML community to crowdsource models and prototypes — a need it addressed by engaging HackerEarth and using the HackerEarth Sprint innovation management and ML platform.

HackerEarth Sprint ran a global ML innovation campaign using HUL’s sales data from India’s top six POS systems, tapping HackerEarth’s developer community to solicit ideas, build models and manage submissions. The campaign attracted 2,304 data scientists and ML enthusiasts, produced 138 idea submissions and 12 winning applications — including week‑ahead sales forecasts to prevent stockouts, a low‑cost Raspberry Pi → S3 analytics pipeline with visualizations, and offline inventory/demand prediction tools — delivering measurable prototypes and improved prediction accuracy for small retailers through HackerEarth’s platform.

Hindustan Unilever Limited