Transcervical endometrial sampling is widely used to study the uterine microbiota, but the risk of cervical contamination raises doubts about whether detected microorganisms genuinely originate from the endometrium. This study aimed to determine whether autochthonous endometrial microbiota can be detected in transcervically collected samples by quantitative PCR after accounting for cervical contamination. Paired cervical and endometrial samples were collected from 185 reproductive-age women using an Endobrush catheter for endometrial sampling. Microbiota composition was assessed by real-time PCR targeting 27 microbial groups. An endometrial signal was considered true if its quantity exceeded that in the paired cervical sample, adjusted for transfer thresholds of 30% (expected, based on prior in vitro data) and 100% (conservative). Before correction, microbial signals were detected in endometrial samples from 144 (77.8%) women. After correction, positive signals remained in 44.3% (30% threshold) and 37.3% (100% threshold) of women. Most Lactobacillus spp. positives fell into an uninterpretable gray zone (47.6% and 55.1% at the 30% and 100% thresholds, respectively). In contrast, opportunistic microorganisms largely remained true positives after correction. Autochthonous endometrial microbiota was detected in approximately 37% of women, typically comprising 1–3 predominantly non-lactobacillus taxa. Detection of opportunistic microorganisms in the endometrium therefore likely reflects their genuine presence, whereas Lactobacillus detection is, in most cases, indistinguishable from cervical contamination.