Educational data’s lifecycle is reimagined in this paper to include the moments that frequently remain in our peripheral vision—as they are little confusing, and a little bit scary. Situated in the classrooms and governing bureaucratic centres of state education in Queensland, Australia, we create a synergetic combination of our two recent research projects. The result is a panoramic view of the life of data from gestation and birth to its zombie half-life. Zombie data is considered data that is no longer needed but remains, also data that you thought had gone, but ‘walks’ again. The paper’s rich empirical data provides the opportunity to present nuanced and novel claims about the life and work of data in education systems. Data were gathered in one study from 52 students from across Years Three to Nine in 19 focus group discussions, interviews with 27 teachers, and observations from within a range of participants’ classrooms are interrogated. Empirical data in the other study were gathered from 68 participants from within the same Queensland state schooling system thereby creating a continuum across all systemic hierarchical levels. Participants engaged in 43 interviews and eight focus group discussions and included school leadership teams in four regional primary schools, as well as senior bureaucrats and technocrats in both Central and Regional offices. Key results show that the lifecycle of data is far broader than is often considered within current literature. The ‘birth’ of data is identified by looking to those who are recognised as the producers. We note that data exist prior to the moment teachers record assessment, behaviour, attendance, and enrolment data, in data infrastructures. The role of those who ‘do’ the assessments, and behaviours, those who arrive at school are identified as the producers, that is, the students. We look to the theory of practice architectures to guide our critical understandings of the sayings, doings and relatings within schools to understand this gestation and birthing of data. Education data are archived according to legislated retention practices long after the students have left school. In this way a student’s Year Two temper tantrum, Year Seven below-national-benchmark numeracy result, and their 15-year-old PISA result eventually die and are forgotten – or do they? We contend that data do not readily die and identify the pernicious ability of digital infrastructures to maintain data’s presence ensuring that they may be brought back to life. Data embalmed through data retention protocols are entombed in archival servers to remain in a state of half-life prior to their final planned destruction. However, these ‘zombie’ data, devoid of their original context or purpose can be summoned by Right to Information requests from media or political lobbyists, sought out in the archival corners of cyberspace to rise again. If data and the use of data are to be fully understood and trusted, there will need to be practice architectures which enable a reframing of not only data’s use but its life and death. We seek to further these discussions in this presentation.