Mapping of the apparent soil electrical conductivity (ECa) can be used to estimate the variability of forage yield within a plot. However, forage production can vary according to the growing season and to soil properties that do not affect the ECa (e.g. nitrogen (N) content). The aim of this study was to assess the relationship between ECa and forage yield of tall fescue (Lolium arundinaceum (Schreb.) Darbysh.) during different regrowth periods and contrasting levels of N availability and then use this information to determine potential management zones. The ECa was measured and geo-referenced in a 5.75-ha paddock that sustained a permanent pasture dominated by tall fescue. In addition, a 30 m by 30 m grid cell size was chosen and 43 sampling areas, each 4 m² in size, were geo-referenced and divided into two experimental units of 1 m by 2 m, one of which was fertilised with 250 kg N ha^–1 (N250) at the beginning of four regrowth periods (spring 2015, spring 2016, autumn 2016 and autumn 2017) and the other was not fertilised with N (N0). At the end of each regrowth period, we estimated the accumulated biomass. During the spring growing season, accumulated biomass was positively associated with ECa in both N0 and N250 treatments (R² = 47% and 54%, respectively). By contrast, in autumn, accumulated biomass and ECa were poorly associated (R² = 10% and 27% for N0 and N250). This may be due to seasonal interactions that alter soil–yield relationships. To assess whether ECa can be used to determine management zones, the differences in accumulated biomass were compared through analysis of variance. Results showed that ECa is associated with the spatial distribution of tall fescue forage yield variability in spring at different N availabilities. Thus, ECa can be reliably used for defining management zones in marginal soils under permanent pastures.