As the complexity of a portfolio grows, sales teams in some markets lean toward defending the business mode rather than actively thinking about customer churn and cross-selling opportunities. This leads to overlooked and unrealized sales. Additionally, for companies with many SKUs, the process of reviewing and individually defining each order quantity results in lost time that could have been spent having more productive discussions with customers to address their needs.
 

ZS partnered with a global healthcare company (focused primarily on generics and OTC) with this exact problem to develop an AI-powered decision engine to increase the efficiency of product recommendation to the customer and uncover cross-selling opportunities.

The company has a strong presence in the OTC and generics space across the EU, and their sales teams work with numerous independent pharmacies. The company has a large portfolio of over 500 SKUs. It recognized the opportunity to leverage AI technology to standardize and automate its sales algorithm across pharmacies to make the order-taking process more efficient and identify cross-selling opportunities.

ZS developed an AI-driven algorithm that can generate the optimal set of order recommendations for each individual customer. The AI-powered “Next Best Order” decision engine leverages the company’s order history, customer segmentation, pricing, customer demographic details and market sales. The decision engine consists of two models:

• A “retention model” that predicts the SKUs and respective quantities each pharmacy will order, based on the transaction history of the past year
• A "cross-sell model” that recommends new SKUs each pharmacy is most likely to purchase, based on product affinity and the pharmacy’s profile

In the testing stages, the retention model successfully predicted orders on par with established decision engines used by supermarket and pharmacy retailers. Field teams validated the cross-sell model to focus on “overlooked” sales opportunities. The engine is powered by multiple algorithms (XGboost, regression, matrix factorization, etc.). The model is designed to learn over time and further improve the relevance of the outcomes. 

The retention model improves the sales team’s quality of working by freeing up bandwidth for value-adding activities. It also prevents sales leakage by reducing bias and manual thinking from the order-taking process, resulting in an overall better customer experience for the pharmacies while also boosting the perception of the company as an innovation-driven organization.
 

The cross-sell model adds value by highlighting unrealized sales opportunities. By encouraging reps to track and convert cross-selling opportunities, it develops a more creative selling mindset in the organization.
 

A business unit head told us: “The cross-selling recommendations are powerful. When we discussed these recommendations with the sales team, they acknowledged that there are some overlooked opportunities. So, we see a lot of promise with this solution and want to integrate it in our day-to-day working.”
 

The engine is expected to generate new selling opportunities. We expect to see a 3%-5% sales uplift. Additionally, we expect to see a 10%-15% reduction in the time it takes a sales rep to visit a client. The client organization is now evaluating expanding the roll-out across the entire EU market.