Ten years after completion of the Human Genome Project, progress towards making "personalized medicine" a reality has been slower than expected. The reason is twofold. Firstly, the science is more difficult than expected. Secondly, limited progress has been made in aligning economic incentives to invest in diagnostics. This paper develops nine case studies of "success" where diagnostic tests are bringing personalized medicine into clinical practice with health and economic impact for patients, healthcare systems, and manufacturers. We focus on the availability of evidence for clinical utility, which is important not only for clinicians but also for payers and budget holders. We find that demonstrating diagnostic clinical utility and the development of economic evidence is currently feasible (i) through drug-diagnostic co-development, and (ii) when the research is sponsored by payers and public bodies. It is less clear whether the diagnostic industry can routinely undertake the work necessary to provide evidence as to the clinical utility and economic value of its products. It would be good public policy to increase the economic incentives to produce evidence of clinical utility: otherwise, opportunities to generate value from personalized medicine-in terms of both cost savings and health gains-may be lost.