Understanding diverse microbial communities is important due to their ecological and medical significance. Bacterial cells are genetically and phenotypically heterogeneous, making their interactions in the communities complex. The heterogeneity and interactions of cells contribute to the formation of specific spatial structures, such as biofilms, and the spread of antibiotic resistance. Here, we describe a novel single-cell approach for studying the cellular heterogeneity and spatial interactions in microbial communities that combines polyacrylamide bead encapsulation of cells and split-pool-barcoding. We demonstrate the method by determining artificially imposed interactions and connecting the taxonomic information in a mock three-species bacterial community with a species-specific genomic target. The method can be utilised for the spatial analysis of microbial communities as well as, once fully optimised for single-cell resolution, linking genetic traits to single cells.