Case Study: a major consumer products manufacturer achieves 98% product identification accuracy with Sigmoid's ML-based image processing

American F100 Consumer Products Manufacturer improves brand detection to 98% with Sigmoid

An American F100 consumer products manufacturer sought to enrich its consumer dataset to gain deeper insights into sentiment, behavior, and purchase patterns. This analysis was crucial for understanding their product placement, validating market share against competitors, and optimizing marketing initiatives. To address this challenge, they partnered with Sigmoid.

Sigmoid developed an app-ready image analytics solution powered by Google Vision API and deep learning. The solution processed consumer shelf images to identify products and brands with high accuracy, using custom NLP and probabilistic matching to eliminate duplicates. Sigmoid's solution achieved 98% accuracy in identifying products and 96% precision in detecting brands. This enabled the client to deeply understand brand penetration, categorize customer segments, and reallocate their marketing budget effectively.