Value-added measures can be used to allocate funding to schools, to identify those institutions in need of special attention and to underpin government guidance on targets. In England, there has been a tendency to include in these measures an ever-greater number of contextualising variables and to develop ever-more complex models that encourage (or "impose") in schools a single uniform method of analysing data, but whose intricacies are not fully understood by practitioners. The competing claims of robustness, usability and accessibility remain unresolved because it is unclear whether the purpose of the measurement is teacher accountability, pupil predictability or school improvement. This paper discusses the provenance and shortcomings of value-added measurement in England (and the Pupil Level Annual Schools Census that informs it) including the fact that although the metrics are essential for School Effectiveness Research, they fail to capture in its entirety the differential effectiveness of schools across the prior attainment range and across sub-groups of students and subjects.