IDENTIFYING & SIZING OPPORTUNITY-NATIONAL LEVEL
Client Situation
Sales at a CPG client were stagnating and underperforming relative to those of main competitors in the category. Client had attempted to reverse the trend through new product introductions but those had only cannibalized the existing offering.
Results
Improved sales by 3% in pilot markets. Identified white spaces to extend brands with $120MM in demand currently unserved by competitors.
SFC Value-Add
We built a shopper segmentation to prioritize which consumers should be targeted with existing brands. We also determined which functional and emotive benefits should be emphasized in new product introductions.
We used a robust ensemble clustering technique to build the segments. To improve the findability of the shopper segments we had created, we combined attitudinal and motivation data from surveys with observable sales and transactional data elements from various sources.
IDENTIFYING & SIZING OPPORTUNITY-LOCAL LEVEL
Client Situation
Client was experiencing unbalanced per capita sales across states. Attempts to lift sales for underperforming regions through advertising or promotional campaigns did not yield any meaningful or long-lasting improvements.
Results
Implemented individual regional growth plans for sales expansion and reversed per capita underperformance in two-thirds of the underperforming states.
SFC Value-Add
To analyze the dynamics of category sales in a robust statistical manner, we built a comprehensive database integrating disparate proprietary and publicly available data sources.
We leveraged multiple Machine Learning methods for estimating the drivers of consumer demand and built financial models for sales expansion. We incorporated cost and distributional assumptions into simulators that the client used to monitor early-warning signs of upcoming drops or surges in demand.
INCREASING PRECISION OF OUTBOUND CUSTOMER TARGETING
Client Situation
A property and casualty insurer had been losing some of its most profitable customers at an alarming rate. None of the broad-based retention initiatives had succeeded in stemming customer attrition.
Results
Improved retention by 7 percentage points and eliminated $1.2MM in wasteful outbound marketing spend.
SFC Value-Add
We built advanced algorithms with Artificial Intelligence components that estimated the individual risk of attrition. We also replaced a static customer scorecards within the client’s CRM with dynamic ones leveraging Bayesian prior-posterior updating.
We designed and maintained a schedule for model scoring in the entire client CRM database of 23 MM records regularly refreshing the risk of defection by incorporating the most recent behavioral and transactional data.
IDENTIFYING LOCATIONS FOR GEOGRAPHIC EXPANSION
Client Situation
A major super market chain sought new store locations with the highest potential for expanding sales to take advantage of the broader market trend of continued expansion in food and beverage sales.
Results
Delivered 1.2% sales increase in existing locations. In addition, identified geographic areas for expansion of new store openings.
SFC Value-Add
We grouped existing stores to prioritize future levels of investment based on current sales, demand profile, and potential for expansion. We devised product shelving, assortment and merchandising guidelines.
We identified geographies for opening new retail locations. To that end we built predictive algorithms, typed all 190MM Households in the US and projected shopper segments to store trading areas using proprietary block group allocations.