The aim of the current study was to investigate to what extent children’s potential for learning is related to their level of cognitive flexibility. Potential for learning was measured through a dynamic testing procedure that aimed to measure how much a child can profit from a training procedure integrated into the testing process, including the amount and type of feedback the children required during this training procedure. The study followed a pre-test–training–post-test control group design. Participants were 153 6–7-year-old children. Half of this group of children were provided with a standardised graduated prompts procedure. The other half of the participants performed a non-inductive cognitive task. Children’s cognitive flexibility was measured through a card sorting test and a test of verbal fluency. Results show that cognitive flexibility was positively related to children’s performance, but only for children in the practice-only condition who received no training. These outcomes suggest that dynamic testing, and more in particular, the graduated prompting procedure, supports children’s cognitive flexibility, thereby giving children with weaker flexibility the opportunity to show more of their cognitive potential as measured through inductive reasoning.