While there has been no shortage of discussion of urban big data, smart cities, and cities as complex systems, there has been less discussion of the implications of big data as a source of individual data for planning and social science research. This study takes advantage of increasingly available land parcel and business establishment data to analyze how the measurement of proximity to urban services or amenities performed in many fields can be impacted by using these data—which can be considered “individual” when compared to aggregated origins or destinations. We use business establishment data across five distinctive US cities: Long Beach, Irvine, and Moreno Valley in California; Milwaukee, Wisconsin; and the New York borough of Staten Island. In these case studies, we show how aggregation error, a previously recognized concern in using census-type data, can be minimized through careful choice of distance measures. Informed by these regions, we provide recommendations for researchers evaluating the potential risks of a measurement strategy that differs from the “gold standard” of network distance from individually measured, point-based origins and destinations. We find limited support for previous hypotheses regarding measurement error based on the abundance or clustering of urban services or amenities, though further research is merited. Importantly, these new data sources reveal vast differences across cities, underscoring how accurate proximity measurement necessitates a critical understanding of the nuances of the urban landscape under investigation as measures appear heavily influenced by a city’s street layouts and historical development trajectories.