With the introduction of Computer Science (CS) into curricula worldwide, researchers have investigated whether CS could be introduced transversally, as a support other disciplines. Few however consider both student learning and the teachers’ perspective in their assessments. In a co-constructive approach to translational research, we collaborated with teachers, in two case studies involving two classes each, to investigate how CS content could be used transversally. More specifically, teacher inputs and student learning data were combined in a mixed methods analysis to determine whether two CS Unplugged (CSU) activities from the curriculum could be leveraged to teach disciplinary content in primary school. The findings indicated that the CSU activities could be leveraged for maths and spelling, but require validation at a larger scale. More important, though, are the takeaways of the co-constructive experience with teachers. Interestingly the benefits of co-construction went both ways. While researchers gained a deeper understanding of the effectiveness of the activities, teachers i) appreciated having detailed student learning analyses, devoid of biases and impressions, ii) changed their perspective about researchers, and iii) were open to other collaborative research initiatives which they found to be drivers of change and innovation in their practice. The way co-construction was approached also impacted the teachers’ perception of their experience. Indeed, while one study was researcher-driven and included teachers, the other was teacher-driven and included researchers. When teacher-driven, the teachers felt they had a more active role, thus reflecting more on the study design, results and implications. Provided the positive impact that co-construction may have on innovation in teacher practices, different co-construction modalities must be investigated, as well as their implications on all stakeholders, including researchers.