Client had an inventory of 3 million SKUs out of which only ~10% had a product hierarchy attached. Categorizing the remaining SKUs was a slow manual process which directly led to loss of revenue and a poor search experience on the website. The task was to automate the process to reduce revenue loss and improve user-experience.
To achieve this, we had to:
30X faster categorization than the existing manual process
99% accuracy for ‘High Confidence’ predictions and overall accuracy of 95%
$250K per year cost reduction in 3rd party expenses for manual categorization
Better search experience on the website and incremental revenue