This record deposits two anonymised datasets together with two codebooks and the full methodological report of the National Congregations Study Switzerland NCSS. It combines (a) two exhaustive censuses of Swiss local religious groups (parishes, mosques, Buddhist centres) conducted in 2008-2009 and 2020-2022 and (b) two waves of a nationally representative key-informant survey implemented in 2008-2009 and 2022-2023. The census file lists every identified Swiss local religious group congregation as a single record and was built using denominational registries, prior inventories, and expertise gathered from religious practitioners and academic specialists. Each observation carries a three-level statut field that indicates its inter-wave trajectory: maintained across both waves, disappeared in between, or newly appeared by 2022. The file further includes, for example, a twelve-category religious-tradition variable, canton, and municipality. The survey waves draw stratified random samples from the census frames, oversampling non-Christian and in 2022 environmentally engaged congregations. Data were collected by YouGov Switzerland (formerly LINK Institute) through computer-assisted telephone interviewing with CAWI follow-up; response rates reached 60.9 % in 2008-2009 and 45.4 % in 2022-2023. Post-stratification and unequal-probability weights correct for non-response and disproportionate sampling allocation. The instrument captures notably organisational characteristics, staffing structure, financial metrics, worship style, youth and educational provision, social composition, environmental action, and member-eligibility norms such as policies on gender and sexual-orientation inclusion. The questionnaire largely replicates the U.S. National Congregations Study, allowing direct international comparison with the American series available via the ARDA repository (https://www.thearda.com/). The dataset includes two weighting variables. coefSample adjusts for unequal selection probabilities and nonresponse, allowing analyses to approximate nationally representative estimates of congregations. w3AttCong weights cases by the number of regular attenders, making it possible to describe the average experience or characteristics of individual congregants rather than congregations.