Improving product profitability by 3% with
ML-based assortment intelligence

Built an ML-based assortment lifecycle solution that helped increase market
share by optimizing investments in the high growth packaged snacks market

Business Challenge

The customer was facing intense competition in the packaged snacks market across its 1100 retail stores due to dynamically changing consumer dietary preferences and price-undercutting of private label products. The vast range of products across various categories, brands, and regions, made it challenging to identify high performing products and develop strategies to increase profitability. The customer wanted to optimize investments towards products with high growth potential into the market while restricting investments where the growth potential was low.

Sigmoid Solution

Sigmoid built an ML solution with best-in-class open source AI/ML libraries such as Pandas, NumPY, to identify product profitability and investment prioritization across its lifecycle. The assortment lifecycle intelligence primarily takes in 3 datasets including Nielsen Scantrack, profitability, and trade promotions data. The solution helped sort products based on growth rate and relative market share of the product considering various internal and external factors such as price and presence in stores to make investment decisions. It also provided a view of the performance of competitor products with that of the brand’s.

Download the Complete
Case Study here

Business Impact

Sigmoid helped the customer align their brand strategy with business rules and promotion guidelines. The customer was able to track movement and lifecycle of products and identify non-performers based on business rules to hold or divest in products accordingly.

3%

Improvement in Contribution Margin

0.8%

Market share growth in test environment

60%

improvement in mapping across Nielsen Scantrack and profitability data

Relevant Blogs

Microservices Architecture Blog Info thumbnail

Microservices-based Architecture: Key to Scaling Enterprise ML Models

ML model training blog thumbnail

Top 5 Model Training and Validation Challenges That Can be Addressed with MLOps

How to Detect and Overcome Model Drift in MLOps Banner image

How to Detect and Overcome Model Drift in MLOps