Introduction The aim of this study was to investigate the differences in shear-wave sonoelastography (SWS) scores between the different parts of cervix, explore the association between the cervical SWS scores with cervical length and evaluate repeatability of the measurement of cervical SWS scores. Material and methods This was a prospective study performed in women with singleton pregnancy at 11-13(+6) (n = 676), 16-20(+6) (n = 364), 21-24(+6) (n = 338) and 28-32(+6) weeks (n = 304). The SWS scores were obtained at the inner, middle and external parts of the cervix, using a transvaginal ultrasound approach. Results The SWS scores of the inner cervix were significantly higher than the measurements acquired at the middle and external parts (all P < .001). At 21-24(+6) and 28-32(+6) weeks, most regions of interest demonstrated a very weak positive correlation with cervical length (r = .125 to r = .299). In comparison with nulliparous women, parous women without prior preterm birth had higher SWS scores of the inner and middle parts of the cervix at 16-20(+6) and 21-24(+6) weeks. All regions of interest showed good intra- and inter-observer agreement. Conclusions The assessment of the cervical SWS scores is highly reproducible. The stiffness of the cervix demonstrates a gradient that decreases from the inner part to the external part and a very weak positive correlation with cervical length.