Myeloproliferative neoplasms (MPNs) are haematopoietic stem cell-derived clonal disorders characterised by proliferation of some or all myeloid lineages, depending on the subtype. MPNs are classically categorized into three disease subgroups; essential thrombocythaemia (ET), polycythaemia vera (PV) and primary myelofibrosis (PMF). The majority (>85%) of patients carry a disease-initiating or driver mutation, the most prevalent occurring in the janus kinase 2 gene (JAK2 V617F), followed by calreticulin (CALR) and myeloproliferative leukaemia virus (MPL) genes. Although these diseases are characterised by shared clinical, pathological and molecular features, one of the most challenging aspects of these disorders is the diverse clinical features which occur in each disease type, with marked variability in risks of disease complications and progression to leukaemia. A remarkable aspect of MPN biology is that the JAK2 V617F mutation, often occurring in the absence of additional mutations, generates a spectrum of phenotypes from asymptomatic ET through to aggressive MF, associated with a poor outcome. The mechanisms promoting MPN heterogeneity remain incompletely understood, but contributing factors are broad and include patient characteristics (gender, age, comorbidities and environmental exposures), additional somatic mutations, target disease-initiating cell, bone marrow microenvironment and germline genetic associations. In this review, we will address these in detail and discuss their role in heterogeneity of MPN disease phenotypes. Tailoring patient management according to the multiple different factors that influence disease phenotype may prove to be the most effective approach to modify the natural history of the disease and ultimately improve outcomes for patients.