Jute technical fibres were tested in tension at 10 different gauge lengths between 6 mm and 300 mm (⩾50 or ⩾100 tests at long or short gauge lengths respectively). The Young’s modulus, strain to failure and ultimate tensile strengths were determined individually and then Weibull distribution parameters were estimated using the maximum likelihood method to quantify the variation. Single Data Set (SDS, standard) and Multiple Data Set (MDS) weak-link scaling (WLS) predictions were assessed using Anderson–Darling Goodness of Fit Numbers (GOFN). The use of MDS predictions provides better correlation with the experimental data than the standard weak-link scaling method. The authors recommend the use of MDS weak-link scaling for this problem with at least two points (but preferably three) with fibre lengths at two extremes (and, if used, the third point near the mean fibre length).