Heart failure (HF) is a threat to public health. Heterogeneities in etiology and phenotype complicate the diagnosis and management of HF. This is especially true when considering HF with preserved ejection fraction (HFpEF) which make up 50% of HF cases. Natriuretic peptides may aid in establishing a working diagnosis in patients suspected of HF, but echocardiography remains the optimal choice for diagnosing HF. Echocardiography provides important prognostic information in both HF with reduced ejection fraction (HFrEF) and HFpEF. Traditionally emphasis has been put on the left ventricular ejection fraction (LVEF). LVEF is useful for both diagnosis and prognosis in HFrEF. However, echocardiography offers more than this single parameter of systolic function, and for optimal risk assessment in HFrEF an echocardiogram evaluating systolic, diastolic, left atrial and right ventricular function is beneficial. In this assessment echocardiographic modalities such as global longitudinal strain (GLS) by two-dimensional speckle-tracking may be useful. LVEF offers little value in HFpEF and is neither helpful for diagnosis nor prognosis. Diastolic function quantified by E/e' and systolic function determined by GLS offer prognostic insight in HFpEF. In HFpEF other parameters of cardiac performance such as left atrial and right ventricular function evaluated by echocardiography also contribute with prognostic information. Hence it is important to consider the entire echocardiogram and not focus solely on systolic function. Future research should focus on combining echocardiographic parameters into risk prediction models to adopt a more personalized approach to prognosis instead of identifying yet another echocardiographic biomarker.