Arid and semi arid environments are characterised by extreme fluctuations in temperature and low rainfall which present significant challenges to the animals inhabiting these areas. Mammals, such as aardvarks (Orycteropus afer, Pallas 1766), excavate burrows in order to avoid predators and climatic extremes and are termed “ecosystem engineers” as they physically modify their environment and in doing so create new habitats and alter the availability of resources to other species. In this study I assessed the microhabitat conditions (maximum and minimum temperature, relative humidity and seed abundance) of aardvark burrows in relation to paired control sites. In addition, I evaluated the use of aardvark burrows by other vertebrate and invertebrate species and investigated the impact of aardvark burrow mounds on landscape scale floristic diversity. Maximum temperatures were significantly lower (p < 0.05) and minimum temperatures and midday humidity were significantly higher (p < 0.05) inside the burrows at the three study sites, Kwandwe Private Game Reserve (Kwandwe), Mountain Zebra National Park (MZNP) and Tswalu Kalahari Reserve (Tswalu). There were no significant differences between the concentration of seeds, the average numbers of unique individual small mammals, trap success or small mammal species richness recorded inside the burrows compared to outside (p > 0.05). At all three sites, small mammal species diversity was higher in the burrows but this result was also not significant (p > 0.05 for all). Trap success and the number of individuals captured was higher at Tswalu than the other two sites (p < 0.05 for both). The different methods used in this study revealed a total of 25 mammal, seven bird, one amphibian and six reptile species utilising aardvark burrows. There were significant differences in insect community assemblages between the burrows and open control areas at Kwandwe and Tswalu (p < 0.05 for both) but not at MZNP (p > 0.05). The parasitic guild was more prominent inside the burrows than outside but their abundance was not as high as anticipated, possibly due to the placement of traps closer to the burrow entrances than the sleeping chambers. The complex structure of the burrows prevented the placement of traps in close proximity to the sleeping chambers. As expected, the amount of bare earth was significantly higher on active and recently abandoned burrow mounds compared to the old burrow mounds and reference plots at all three sites (p < 0.05 for all), with the exception of the active burrows at Tswalu. Overall, the different plot types were characterised by significantly different plant communities during all the seasons at MZNP, during three of the seasons at Kwandwe and only during winter at Tswalu. The total species richness recorded on the reference plots was higher than on the burrow mounds at all three sites. However, species diversity on the reference plots was not significantly higher than the burrows at any of the sites (p > 0.05 for all sites). Although the results were not significant, the overall species diversity at a site level was greater than the reference patches at Kwandwe and Tswalu (p > 0.05 for both). Aardvarks fulfil the criteria of a significant ecosystem engineer and their presence in arid and semi-arid environments is likely to be critical to the survival of other individual organisms and species, particularly when alternative burrowing animals are either absent or restricted in their activities. Thus, aardvark populations should be considered a conservation priority in arid and semi-arid ecosystems.