Leveraging harmonized data for out-of-stock detection and powering ML applications

Out-of-stock is a common problem for personal care manufacturers, as it affects their in-store presence and sales performance. Detecting and preventing out-of-stock requires a reliable data source, in-depth understanding of the industry and specific ML/ AI tools for actionable results.

Our client, an industry-leading personal care manufacturer, wanted to automate their out-of-stock detection. They had an IT department that was capable of developing and testing algorithms, but they needed a harmonized and enriched data source that could feed their models.

That’s where Redslim came in. We provided the client with harmonized data from more than 40 countries and 4 categories, in a format that was compatible with their machine learning systems. We also added an enrichment layer that included additional information such as store location, product attributes, and shelf layout.

Our data ensured that client’s data models were reliable and accurate, as they were based on high-quality, harmonized, and clean data that eliminated inconsistencies and errors. With our data, the client was able to run their algorithms and identify key issues with their stock across a whole region. They were able to give recommendations to their local business units on how to improve their supply chain and inventory management, and ultimately increase their sales.

The project was completed in a few weeks, and resulted in a successful implementation for out-of-stock detection. Today, the client continues to use our data and services to monitor and optimize their in-store performance.

Solution

Redslim provided the client with harmonized and enriched data from multiple countries and categories, which enabled them to automate their out-of-stock detection and improve their supply chain and sales.