Polymicrobial bacterial infection is an important factor contributing to wound chronicity. Consequently, clinicians frequently adopt a biofilm-based wound care approach, in which wounds are treated utilizing DNA sequencing information about microbial communities. While more successful than treatment not using community information, there is little information about temporal dynamics of wound communities and optimal approaches over the course of treatment. To characterize these dynamics, temporal analysis over three sampling points was conducted for 167 chronic wounds. Across sampling intervals, wound communities from the same patients changed in composition, and most commonly shared less than 50% of observed species. There was a significant relationship between community similarity and time between sampling. Classifying wounds into state types, we found that communities frequently transitioned from Pseudomonas or Staphylococcus dominated, into a highly variable state type. Although low abundance microbial species are typically disregarded due to uncertainty of biological importance, we found that 80% of wound microbiomes included common or dominant species at subsequent time points that were in low abundance in earlier samples. Moreover, these species were often those known to frequently infect wounds. Results document compositional shifts through the course of treatment and suggest that routine consideration of low abundance species may improve biofilm-based wound care. Moreover, findings indicate that integrating ecological modeling to understand wound microbiome succession may lead to more informed therapy.