One important caveat discussed in Euclid Collaboration: Le Brun et al.(2025) is to be careful when using redshifts from the catalogues. On the one hand, by design, the SGS pipelines produce one spectrum for each source of the MER catalogue with . This corresponds to about 4.3 million spectra. On the other hand, the expected number of sources with H flux above the nominal flux limit of the EWS (predicted from luminosity functions for an area of 63.1 deg2) is of the order of 100 000. This corresponds to about 2.5% of the spectra. These data also include further spectra for stars and galaxies/AGN outside of the target redshift range, measured from other lines. These will still be a small part of the total, so that the fraction of spectra allowing for the measurement of a redshift is below 10%. Yet, the SGS pipeline provides a redshift estimate for everyone of them. The challenge the SGS is confronted with is to correctly assess the reliability of the automated redshift measurements. The issue will be even more complex for DR1 when attempting to increase the limiting magnitude of the extracted sample. Meanwhile, we recommend to work with the highest quality of spectra, and to check if possible the redshift by examining the 1D spectra of the sources of interest. This can only be done for small samples of sources.