Albeit endometriosis is one of the most common gynecological diseases, its diagnosis and treatment remain controversial. The reasons behind this include: 1) multifactorial pathogenesis and insufficiently studied mechanisms of endometriosis; 2) relatively low diagnostic value of minimally invasive examination in relation to this disease; 3) inefficiency of current therapeutic approaches in many patient settings. In our opinion, uncovering the causes of endometriosis and factors promoting its progression is the cornerstone of its successful management. Here we review the lessons from genome-wide and candidate gene association studies, discuss the expression of regulatory miRNAs and describe the role of heat shock protein 90, annexin A2, and peroxiredoxin 2 in controlling DNA integrity in the eutopic endometrium. Further, we highlight the role of cytokeratin-19 in urine as a feasible diagnostic marker of endometriosis. Clinicians and basic researchers concur that the molecular basis of endometriosis is still in its infancy and current understanding of its pathophysiology remains poor. Recent progress in -omics approaches and bioinformatics paved the way for complex investigations of regulated cell death, proliferation, cell invasion and angiogenesis, opening the avenue for the novel approaches to treat endometriosis. Yet, the diversity of symptoms and an absence of sensitive and specific biomarkers frequently delay and complicate the diagnosis. In addition, surgery represents the only appropriate option to reliably confirm the diagnosis and to establish the disease extent, reducing patient adherence and postponing the start of the treatment. In this review, we discuss challenges in the diagnosis of endometriosis as well as relevant and potentially informative biomarkers.