The Panel 2008 is part of a longstanding tradition of panel surveys conducted by the DARES since the early 1990s and aims to describe the career paths of beneficiaries of subsidised labour contracts (CIE, CAE, CAV, and CI-RMA) and to evaluate the effect of the switch to subsidised contracts on these paths. Indeed the Panel 2008 was designed with evaluation in mind, on the basis of a comparison between beneficiaries and eligible non-beneficiaries using, among other things, matching techniques. The survey was conducted in two waves at the end of 2008 and the end of 2009, on a sample of jobseekers who had registered with the ANPE in spring 2005, some of whom were hired on a subsidised contract between Q2 2005 and Q2 2007 while the others were not. Comparability between beneficiaries and non-beneficiaries is a crucial assumption of matching techniques. The protocol for the Panel 2008 sought to guarantee this assumption as early as the sampling stage, which was carried out on a single source, but also in designing the questionnaires, which were devised to be as similar as possible to each other for the two populations, as well as in the processing of the data collected (correction for nonresponse, and calibration). Similarly, the survey includes numerous variables serving to provide a detailed measurement of the unobserved heterogeneity of respondents, particularly in terms of new dimensions such as their relationship with studies, their literacy and their state of health. The richness of data available in the questionnaire also allows a causal effect to be identified between the switch to a subsidised contract and several result variables, such as the likelihood of being in employment at certain dates or changes in the living conditions of beneficiaries. This article proposes a detailed presentation of the design and conduct of this original survey, one which is of importance in the evaluation of government policy on employment.