Date of Graduation

5-2026

Document Type

Thesis

Degree Name

Bachelor of Science in Data Science

Degree Level

Undergraduate

Department

Data Science

Advisor/Mentor

Karl Schubert

Committee Member

Jacob Yates

Second Committee Member

Eric Specking

Abstract

When companies acquire beverage brands, they typically value them based on total sales revenue. This traditional approach treats all sales equally over time, whether they are driven by genuine consumer demand or temporary discounts. This is important because while promotions can boost short-term sales, they tend to erode brand value over long periods of time. The measurement problem extends to acquisitions, where buyers lack the tools to distinguish real consumer demand from artificial promotional inflation.

This thesis develops a framework to separate genuine baseline demand from promotional dependence using Nielsen scanner data covering 189 beverage brands across 188,304 weekly observations across three U.S. regional markets. The analysis reveals that many brands analyzed suffer from zero or near zero baseline demand and exist solely through promotional activity. Despite generating substantial revenue, these brands lack a sustainable amount of consumer loyalty and can be high risk for acquisition. The beverage market is highly promotional dependent, and the median brand derives approximately 65% of volume from promotions, confirming that promotional dependence is widespread across the beverage category.

The methodology employs three complementary approaches for creating a framework of analysis. First, demand decomposition quantifies how much of each brand's sales come from standard consumer purchases versus promotional discounts. By using metrics such as promotional dependency and baseline market share, we can better understand the market. Second, econometric analysis estimates price elasticity through regression modeling to measure how much certain brands retain their pricing power or have trained consumers to strictly buy on discounts. Third, machine learning techniques segment brands into archetypes and enable predictive quality screening. K-means clustering identifies five distinct brand types based on metrics such as baseline demand. A supervised learning framework comparing Random Forest, Gradient Boosting, and Logistic Regression can be complementary to predict a brand's overall quality. Finally, feature importance analysis reveals that baseline demand strength is the primary predictor of brand quality, followed by promotional dependency, confirming that demand structure drives long-term value.

The resulting framework provides a data-driven screening tool to avoid value-destroying deals for acquisitions. The system includes a demand quality assessment that separates sustainable consumer pull from promotional inflation, a brand attractiveness classification that categorizes acquisition targets, a promotion-adjusted valuation approach, and archetype-specific integration guidance. By incorporating demand decomposition, econometric analysis, and machine learning, acquirers can systematically identify brands with durable competitive advantages while avoiding overpayment for temporary promotional volume.

Keywords

Machine Learning; Consumer Packaged Goods; Beverage; Mergers & Acquisitions; Technical Framework; Promotions

Download

Share

COinS