Ecological studies on native bees in urban tropical environments are rare, however, ever-increasing urban areas world-wide necessitate study on how many of these bees can and have adjusted to human constructs. Predictable ecological patterns that emerge from these studies can provide guidance on how future urban constructs can be designed to provide habitat for conserving and protecting native bees. These patterns can also be used for bee habitat restoration in natural and agricultural environments. An extensive survey of native bees and honey bees and their relationships to a community of 102 plant types in urban residential environments of Bagaces and Liberia in northwestern Costa Rica was conducted from 2003–2012. Bees were attracted and recorded at measurable frequencies to 82 plant genera in 41 families, the most common of which was Fabaceae. Forty-two plant types were native ornamentals; 39 were non-natives; and 21 were native weed species. Standardized bee visitation (frequency) counts, 17,000 , were used to record relationships between bees and flowers. The following data were recorded for each plant type: flowering phenology in months, type of floral reward(s) (pollen, nectar, and/or oil), main daily attraction period, and most frequently visiting bee taxa. Plant life forms included trees, shrubs, lianas/vines, herbs, and palms. Each plant group had a different seasonal flowering phenology with native ornamentals and native weeds having patterns that closely resembled the general patterns for wild plants in the dry forest. Predictable associations of certain bee taxa with each plant type emerged from the count data, which allowed for categorizing relationships into four types: small bee, diverse bee, specialized bee, and nocturnal pollination systems. Intraspecific variations in bee attraction to several plant types were also noted. Honey bees (Africanized) did not figure prominently in most pollination relationships, especially with regard to native plants. Most native bee species were generalized foragers. Beyond the urban environment, it is suggested that knowledge of predictable bee-flower relationships can also be used to restore bee habitat in disturbed environments such as deforested areas. With some imagination and outreach education, bee habitats could also be installed for some agricultural crops. Outreaching information on native bee-flower relationships at local, regional, and state levels is important for short and long-term propagation of urban (and agricultural) plants. Yet, very few outlets for transferring this knowledge currently exist in Costa Rica. A few limited options for sharing this information are discussed, including collaborative partnerships with local NGOs.