Effective sampling is essential for exploring microbial communities. Unfortunately, some microbiomes are harder to explore than others. Samples gathered from the skin, placenta, and lungs tend to be microbially sparse, making the detection of microbial DNA difficult and likely to be hampered by contamination. Now, new research points to at least one way to improve microbial sampling in the lungs of mice. Researchers compared bacterial DNA collected from two different sources: homogenized whole lung tissue and bronchoalveolar lavage (BAL) fluid. Whole lung tissue proved the more ideal source. Compared with BAL fluid and negative controls, whole lung tissue contained more bacterial DNA, had a more distinct microbial composition, showed lower sample-to-sample variation, and contained more biologically plausible bacterial taxa found in the mouth. While the findings are limited to the sampling methods compared in this study and apply only to results gathered by amplicon-based sequencing, they demonstrate the power of using ecology-based analytical approaches and the importance of sequencing, analyzing, and reporting ample negative controls to reflect the biological reality of microbially sparse tissues.