Due to the decrease in fluid milk consumption as a beverage in the United States, the importance for dairy industry stakeholders to understand attribute preferences for consumer segments has increased. This paper uses a representative sample of U.S. residents to determine shopping behavior and a Best-Worst experimental design to examine consumer preferences for select milk attributes. The Random Parameters Logit model revealed the largest preference shares were for price (19.1%), fat content (17.6%), and humane handling (16.1%). Segmentation of the respondents was analyzed using a Latent Class Model, and the demographics of segments were analyzed by probabilistic assignment. This contribution, which may be applied to other products, allows for a detailed understanding of consumer preferences for fluid milk. Demographics such as gender and age were not statistically different across the five classes for this topic, unlike shopping behavior. Class 2 was named “The Balancing Act” due to respondents’ balance between animal welfare aspects and the physical characteristics of milk. Class 5 was dubbed “Value and Volume” due to large preference shares for price and container size. A lower percentage of respondents in “The Balancing Act” (12.2%) reported purchasing fat-free skim milk when compared to the percentage of respondents in “Value and Volume” (18.6%). Due to the large preference shares for traditional milk attributes, Class 3 was named “Traditional Milk Shoppers.” A higher percentage of respondents in “Value and Volume” also reported always reading the information on meat, egg, or milk products when compared to “Traditional Milk Shoppers” (12.2%). Although price is still important to many consumers, certain segments have preferences or demand for other attributes that may be satisfied by producers to increase market share or price premiums.