Chronic obstructive pulmonary disease (COPD) is defined based on a reduced ratio of forced expiratory volume in one second (FEV1 ) to forced vital capacity (FVC) on spirometry. However, within this definition, there is significant heterogeneity of pathophysiological processes that lead to airflow obstruction and variation in phenotypic manifestations across patients. Current pharmacological treatments are based on large randomised clinical trials that apply to an "average" patient. Precision health enables tailoring of treatment for each individual patient by taking into account their unique characteristics. The number needed to treat (NNT) metric is often used to define implementation of precision health for specific interventions, with common endpoints requiring an NNT ≤5 to achieve precision therapy. Higher NNTs may be acceptable for rare but important endpoints such as mortality. Long-acting muscarinic antagonists and inhaled corticosteroids, which are commonly used in COPD, have 1-year treatment NNTs between 15 and 20 for exacerbation prevention in unselected patients with COPD. Subgroup identification using biomarkers or clinical traits may enable precision health. For example, NNT for inhaled corticosteroids is 9 in patients with a blood eosinophil count ≥300cells/μL and 8 for long-acting muscarinic antagonists in patients with a body mass index ≤20kg/m2 . Lung volume reduction surgery is associated with an NNT of 6 for survival over 5 years in patients with upper lobe-predominant disease and low exercise capacity (whereas the NNT is 245 when no bioimaging or exercise markers are used). Continuous domiciliary oxygen therapy (for at least 15hours/day) has an NNT of 5 for survival over 5 years in patients with resting hypoxemia (PaO2 <60mmHg on room air). Emerging areas of precision health in COPD with potential for low NNTs in specific circumstances include anti-interleukin-5 therapy for eosinophilic COPD, and immunoglobulin replacement therapy for patients with severe immunoglobulin deficiency.