The  appeal  of  audiovisual  products  comes  from  the  combination  of visual   and   auditory   elements, but   professional   subtitling   remains focused  on  words,  reducing  other  elements  to  a  contextualising  role. This  assumes  that  nonverbal  elements  such  as  images  or  sounds  are universal codes easily interpreted by viewers without further mediation and  potentially  leaves  some  viewers  with  glaring  losses  of  meaning (Cavaliere, 2008). In this article, I contend that all elements co-occurring with  speech  are  signs  in  their  own  right  that  might  present  different challenges  to  (different)  viewers  and,as  a  result,  might  need  to  be translated.  The  article  reports  on  an  exploratory  experimental  study focused on comparing the impact on viewers’ meaning-making  of a)current subtitling practice focused on verbal signs, and b)innovative subtitling   practice   aiming   at   translating   meaning   expressed   by nonverbal  elements  identified  as  cultural-specific.  The  results  point towards  the  need  for  a  fundamental  shift  in  our  understanding  of nonverbal elements and the need to translate them.