Noise contributions to a GGL tangential shear correlation function measurement for BOSS lenses in the KiDS-1000 overlap (L1) and KiDS high-redshift sources (S5). Shown is the standard deviation as a function of angular separation, normalised by the analytic expectation for uniformly distributed galaxies and neglecting survey boundaries (blue). The noise term as implemented by Eq. (E.3) is shown in orange. It is in excellent agreement with the sample variance obtained from a measurement with randomised source galaxy shapes and lens galaxy positions, i.e. removing all sample variance contributions (purple). Decreasing the oversampling factor of the random catalogue from our default of 𝒩_rnd = 100 to only five results in an increased noise level (brown). We also show the signals when only randomising shapes (red) and when additionally not subtracting the GGL signal around random points in the lens sample (green).