Detecting and characterising exoworlds around very young stars (age$<$10 Myr) are key aspects of exoplanet demographic studies, especially for understanding the mechanisms and timescales of planet formation and migration. However, detection using the radial velocity method alone can be very challenging, since the amplitude of the signals due to magnetic activity of such stars can be orders of magnitude larger than those induced even by massive planets. We observed the very young ($sim$2 Myr) and very active star V830 Tau with the HARPS-N spectrograph to independently confirm and characterise the previously reported hot Jupiter V830 Tau b ($K_{rm b}=68pm11$ m/s; $m_{rm b}sini_{rm b}=0.57pm0.10$ $M_{jup}$; $P_{rm b}=4.927pm0.008$ d). Due to the observed $sim$1 km/s radial velocity scatter clearly attributable to V830 Taus magnetic activity, we analysed radial velocities extracted with different pipelines and modelled them using several state-of-the-art tools. We devised injection-recovery simulations to support our results and characterise our detection limits. The analysis of the radial velocities was aided by using simultaneous photometric and spectroscopic diagnostics. Despite the high quality of our HARPS-N data and the diversity of tests we performed, we could not detect the planet V830 Tau b in our data and confirm its existence. Our simulations show that a statistically-significant detection of the claimed planetary Doppler signal is very challenging. Much as it is important to continue Doppler searches for planets around young stars, utmost care must be taken in the attempt to overcome the technical difficulties to be faced in order to achieve their detection and characterisation. This point must be kept in mind when assessing their occurrence rate, formation mechanisms and migration pathways, especially without evidence of their existence from photometric transits.