Our aim was to assess the effect of a self-correction spelling tool based on orthographic revision procedures on third-grade children's orthographic performance. This tool consisted of grids displaying explicit contextual, phonological and morphological rules. Participants were 70 third-grade students, randomly assigned to an experimental (N = 35) and control group (N = 35). A pretest and post-test were performed a week before the training programme and two weeks after the end of it, consisting of a 76-word spelling test that referred to contextual (32 words), phonological (32 words) and morphological (12 words) rules. The experimental training programme was carried out with children in class and consisted of nine sessions in which children had to spell words dictated by an adult, who then underlined their misspellings with a specific colour (a different colour was assigned to each rule) and asked them to self-correct the misspellings using the grid. None of the words present in the pre- and post-test were used in the training programme. The same words were used in the control group, but instead of self-correcting the misspellings, children were shown the correct spelling and had to copy it three times. The post-test results showed that the experimental group decreased significantly the number of misspellings when compared with the control group, with statistically significant differences for each rule. Nevertheless, the morphological misspellings, after the intervention, were superior to the other types of misspellings. These results show the importance of inducing self-correction spelling strategies in educational contexts.