In designing the city of the future, city managers and urban planners are driven by specific citizens’ behaviors. In fact, economic and financial behaviors, and specifically, which goods and services citizens purchase and how they allocate their spending, are playing a central role in planning targeted services. In this context, cashless payments provide an invaluable data source to identify such spending behaviors. In this work, we propose a methodology to extract the consumption behaviors of a large sample of customers through credit card transaction data. The main outcome of the methodology is a concise representation of the economic behavior of people residing in a city, the so-called city consumption profile. We inferred the city consumption profile from a network-based representation of the similarity among the customers in terms of purchase allocation; on top of which we applied a community detection algorithm to identify the representative consumption profiles. By applying the above methodology to a set of credit card transactions of an Italian financial group, we showed that cities, even geographically close, exhibit different profiles which makes them unique. Specifically, usage patterns focused on a single type of good/service—mono-categorical consumption profile—are the main factors leading to the differences in the city profiles. Our analysis also showed that there is a group of consumption profiles common to all cities, made up by purchases of primary goods/services, such as food or clothing. In general, the city consumption profile represents a tool for understanding the economic behaviors of the citizens and for comparing different cities. Moreover, city planners and managers may use it in the outline of city services tailored to the citizens’ needs.