The COVID 19 pandemic illuminates the role data has in public policy-making, i.e. datafication of society, and the importance of exploring the local sources of data to reveal errors in what has assuredly been from the beginning an undercount of cases and deaths. I note four interrelated error sources. The first two are common to any quantitative data collection project: (1) representation, measurement, and data processing; and (2) problems of data standardization from unequally resourced local and national data providers. COVID 19 casts a special light on (3) the possibility of government intervention in at least the public presentation of these data; and (4) human errors in the data chain caused by a stressful data collection environment. To identify errors, we should look to national pressures and the local contexts from which these data are collected and the upstream reporting process.