In sociophonetic studies, the goal of eliciting natural speech is often at odds with the goal of eliciting speech that is easily comparable across participants. In this examination of regional pronunciation variation among English-speaking children and adults in the United States, these goals were balanced using two speech elicitation methods: color naming and picture-prompted storytelling. In the color naming task, participants saw solid blocks of color on a computer screen and were recorded saying the name of each color. In the storytelling task, participants were recorded narrating the familiar stories of Little Red Riding Hood and Goldilocks and the Three Bears using self-paced picture prompts. These methods were successfully used with participants of a wide range of ages, from pre-literate children (4 years old) to older adults (75 + years old). The color naming task yielded largely identical target words across participants because only common colors were presented. While the storytelling task allowed participants to be more creative, targets such as character names, words for important objects in the stories, and canonical lines of dialog were repeated frequently across participants. Thus, these elicitation methods avoided naturalness concerns associated with read speech while providing identical word-length and similar sentence-length examples for comparison.