The aim is to examine the impact of interventions on fluency and reading comprehension and how the effects of these interventions depend on the time that teachers spend with children with reading difficulties. Two groups were involved: an experimental group (n = 600) trained in code-related skills and a control group (n = 597) that received no intervention. In the Exp group, teachers adapted a specific instructional time for reading (SITR) in the light of children’s difficulties. A significant gain of .23 σ in fluency and a gain of .33 σ in reading comprehension were observed in favor of the Exp group, (a) but there was no gain in numeracy skills (as expected); (b) the amount of SITR is linked to children’s scores in literacy skills; (c) SITR is linked to progress in fluency and reading comprehension. These results argue in favor of differentiated reading instruction for children with difficulties during learning to read.