SMS helps clients maximize profitability and manage risk. Our proprietary, predictive modeling and testing solutions quantify the impact of price, promotion, advertising, online media and assortment on brand and retail performance.
We integrate point-of-sale demand modeling with survey data analytics to provide a complete understanding of the factors that drive consumer demand, needs, attitudes, and shopping behaviors.
Modeling Architecture
SMS modeling systems are built on four leading statistical platforms:
- Bayesian Hierarchical (BH) regression enables SMS to generate stable, multi-level nested estimates with limited data.
- Multiple Comparison Procedures (MCP) drive our experimental designs where multiple test conditions must be simultaneously analyzed to understand the difference between marketing programs.
- Conjoint, Logit & Discrete Choice Models (CL-DCM) quantify the determinants of consumer choice between two or more mutually exclusive alternatives.
- Seemingly Unrelated Regression Equations (SURE) form the core of SMS competitive interaction and attribute measurement models, consisting of multiple regression equations, each with their own dependent and exogenous explanatory variables.
Data Sources
SMS models are applied to a significant range of data sources to ensure a complete view of our clients’ business:
- Point-of-Sale: Retail Link, Food, Drug, Mass, CE & DIY
- Third Party: Nielsen, IRI, NPD, TNS, Dunnhumby
- Household: Consumer, Panel, Survey & Loyalty
- Digital: Social Media, Search, Banner, Website, Comscore, Forrester
- Other: Client Shipment, Agency & Vendor
