Retail Use Cases

Supply Chain Network Optimization

Global Industrial Company: This company had a very extensive global supply network where suppliers dealt with divisions and where supplier performance levels and lead-times differed from division to division around the world.  The company was looking for a way to understand and explore all of the different decision alternatives and make those available to the decision makers __ enabling them to make decisions that were more globally optimized versus restricted to a division.

It was possible to gain a global view of the supply network in terms of supplier performance. SMART AI was used on internal data records to consider scenarios to reduce costs globally.  Network modeling was used to develop and understand more decision alternatives and enabled decision makers to optimize their supplier performance from a global perspective rather than a more narrow, local perspective. 

Result: 20% reductions in inventory levels and logistics costs

End-to-End Inventory Optimization

Global Beverage Company: Responsible for sourcing and managing beverage ingredients globally, this company was challenged to optimize its inventory size and composition per location to reduce waste and storage costs.  They asked for assistance in optimizing their stocking strategy for every SKU per storage location.  Using analytics, their organizational focus on local storage optimization was shifted to a global, end-to-end storage optimization model by consolidating disparate datasets and automating their routine analysis.

Result: 25% reduction in inventory levels without affecting revenue.

Customer Segmentation

Customer-Lifecycle Prediction

Global fashion company: This global fashion company wanted to improve the effectiveness of its annual marketing plan and its execution.  They did not have tools to precisely predict customer behavior and data problems prevented decision makers from developing a strategy to introduce personalization features to their customer experience.

Using an advanced analytics approach, a customer-lifecycle value microapp was developed that was able to provide accurate predictions of when each customer was likely to make his/her next purchase and how many purchases a given customer will make in a specified sales period. With such a high prediction accuracy, decision makers were able to create optimized sales strategies to increase revenue.

Results: 99% average prediction accuracy achieved


E-Mail Marketing and

Cross-Sell Optimization

Global fashion company: This fashion company needed to improve its online sales but was struggling to launch a productive targeted email campaign to communicate with customers most likely to purchase and present suggested additional items that each customer had a high likelihood of purchasing.  Decision makers did not have an understanding of the relevant commercial and sales performance drivers that were affecting their conversion rates and online sales.

An advanced analytic application was developed to optimize customer sales recommendations and extract insights from customer browsing history.  The accuracy of the predicted customer behaviors was enhanced by using external datasets relevant to commercial trends __ thus providing context to customer behavior improving conversion rates tremendously.

Results: 8.4% higher conversion rate with the targeted e-mail campaign and 15.4% increase in revenue.


Price Optimization

Online CPG: A major retailer that offered consumer products online was challenged with ineffective pricing strategies and techniques.  The company needed to understand the demand drivers, monitor the marketplace effectively, and develop a pricing strategy to increase profits.

Working with the ‘pricing teams’, it was possible to use SMART AI to expand from a handful of pricing strategy alternatives to a multitude of alternatives that were defined systematically in order to consider the full scope of options.  In addition, the demand prediction of their online products improved dramatically allowing foresight to avoid ‘stock-outs’.  

Result: 15% higher revenues and improved customer satisfaction by avoiding ‘stock-outs’

Promotion Optimization

Global Quick Service Restaurant Chain: A global QSR chain in need of improvement of their promotional budget spending strategy.  They wanted to optimize their advertising spend related to GRP acquisition and coupon-based discounts. 

For coupon-based discount optimization, the timing and dollar amount of discount allowed per item had to be calculated for a multitude of scenarios so that decision makers could optimize their strategies. Without curated datasets and machine learning, this would not be possible to the level of accuracy that was achieved.

Result: 98% daily prediction accuracy on promotion revenue.


Price Optimization

Global Personal Care Product Company: This international, personal-care brand runs promotions in multiple channels for different product categories on an annual basis. It wanted to understand the impact of price differences with its competitors on its market share for a few of its most competitive products. 

We  established the basis for a multi-dimensional pricing model by understanding how market dynamics were affected by price changes. We called this a price-demand sensitivity analysis. 

In addition, we developed a microapp to deliver optimized predictions and prescribed actions for decision-makers. The microapp simulated market variations resulting from price changes, allowing for “what-if” scenario analysis, and then optimization analysis—in this case on price related to market share. The app was delivered to the company via a web-based interface.

Sales Force Productivity and Optimization

An international pharmaceutical company was having difficulty with its sales force productivity and were looking for changes that could be made to improve their market coverage strategy and product offerings to improve the success of their sales force.

Analysis of purchase history data from doctors and pharmacies along with external data providing context to the purchase history, allowed for an optimized customer segmentation plan of the various markets.  The sales force better understood their customers in terms of when, where and what to sell to each customer segment.

In addition, new employee incentives sales programs were optimized and outlined for the sales force by applying SMART AI to a combination of data leading to __ restructuring of incentives for their teams that improved sales and benefits to the teams.

Result: 15% increase in sales revenue.