Relative performance evaluation (RPE) is a form of benchmarking that operates through ranking institutions in comparative league tables. This paper explores issues raised by the introduction of RPE to benchmark UK hospital costs (termed 'reference' costs for this purpose). These reference costs are aggregated into a comparative cost index-thereby creating a 'ladder of success'.The 'ladder of success' has the potentiality both to enhance the purchasing role and to provide a comparative database for hospitals to improve their performances. Yet this research found that several problems confound its use for benchmarking: the absence of a referent (or standard) against which 'reference' costs can be compared; the non-comparability of many hospitals featuring in the index; and the lack of standardisation in costing practices.In terms of refining and developing the index to enhance its usefulness, there are several possibilities: distinguishing between direct and indirect costs; introducing benchmarking 'cluster groups'; pinpointing an acceptable range of target costs; and reducing the scope of clinical activities included. However, if any of these alternatives were adopted, the power of a single comprehensive measure that attaches one-and only one-number to each trust would be lost. The issue is that in making the index more meaningful, political leverage over the UK trusts would be reduced. Hence, despite all its associated problems, this research concludes that the 'ladder of success' looks likely to continue in its present form.