Problem solving and researching are connected activities in the engineering profession and across education. The empirical research presented in this paper considered the effectiveness for teaching and learning of a model that was based on this connection and derived from the parameters of the Research Skill Development framework. The model, devised by students for students, is called the Optimising Problem Solving (OPS) pentagon, and was piloted in a large first-year engineering course in 2014-2015 and in another university in the first half of 2016. This article presents data from formal research conducted on the 2016 offering of the course gathered from pre (n=221) and post (n=169) surveys, and semi-structured interviews of students (n=5) and student/tutors (n=5) sixteen months after course completion. Statistically significant (p < 0.01) changes from pre to post in items with medium to large effect sizes provided student perspectives on which problem-solving skills improved, and the interview data provided rich detail on the nature of changes and how these changes were effected. For the 2016 cohort studied, our evidence suggests that use of the OPS pentagon was associated with improved student problem-solving skills, and tutors’ capacity to teach problem solving.