Background: In adult patients with bronchiectasis (BE) the identification of the underlying aetiology may be difficult. In a new patient with BE the performance of a panel of tests is recommended, even though this practice may be expensive and the level of evidence supporting is low. We aimed to identify a panel of variables able to predict the aetiological diagnosis of BE. Methods: Our prospective study derived from our real-life experience on the management of adult stable BE outpatients. We recorded variables concerning clinical, radiological, microbiological and laboratory features. We identified five groups of aetiological diagnosis of BE (idiopathic, post-infective, COPD, asthma and non-common diseases [immunodeficiency or other rare conditions]). Multivariate models were used to identify predictors of each aetiological diagnosis. The suitability of performing a specific test for the diagnosis was also considered. Results: We enrolled 354 patients with a new diagnosis of BE. Patients with different aetiological causes differed significantly with regard to age, sex, smoking habit, comorbidities, dyspnoea perception, airflow obstruction and severity scores. Various predictors were assessed, including sex, previous respiratory infections, diffuse localization of BE, risk scores, and laboratory variables (sodium and eosinophils). The levels of autoantibodies or immunoglobulins were reserved for the diagnosis of non-common disease. Conclusion: Our research confirms that some predictors are specific for the aetiological diagnosis of BE. The possibility of integrating this information may represent a useful tool for the diagnosis. The execution of certain specific tests should be reserved for patients with a non-common disease.