Treatment failure (TF) is defined as a clinical condition with inadequate response to antimicrobial therapy. Clinical response should be evaluated within the first 72 h of treatment, whereas infiltrate images may take up to 6 weeks to resolve. Early failure is considered when ventilatory support and/or septic shock appear within the first 72 h. The incidence of treatment failure in community-acquired pneumonia is 10 to 15%, and the mortality is increased nearly fivefold. Resistant and unusual microorganisms and noninfectious causes are responsible for TF. Risk factors are related to the initial severity of the disease, the presence of comorbidity, the microorganism involved, and the antimicrobial treatment implemented. Characteristics of patients and factors related to inflammatory response have been associated with delayed resolution and poor prognosis. The diagnostic approach to TF depends on the degree of clinical impact, host factors, and the possible cause. Initial reevaluation should include a confirmation of the diagnosis of pneumonia, noninvasive microbiological samples, and new radiographic studies. A conservative approach of clinical monitoring and serial radiographs may be recommended in elderly patients with comorbid conditions that justify a delayed response. Invasive studies with bronchoscopy to obtain protected brush specimen and BAL are indicated in the presence of clinical deterioration or failure to stabilize. BAL processing should include the study of cell patterns to rule out other noninfectious diseases and complete microbiological studies. The diagnostic yield of imaging procedures with noninvasive and invasive samples is up to 70%. After obtaining microbiological samples, an empirical change in antibiotic therapy is required to cover a wider microbial spectrum.