This study addresses a gap in the empirical research about how students use computational thinking (CT) skills throughout the problem-solving process, which is a crucial skill for STEM learning. I conducted task-based interviews with two pairs of Year 8 students with strong mathematical problem-solving skills and previous experience in solving mathematical problems using spreadsheets. Data were gathered through audio recordings, computer screen recordings, and copies of student work, and analysed using verbal protocol analysis. The students solved an area optimisation problem, and the findings showed that they used abstraction and decomposition mainly during the Analysis stage, and algorithmic thinking, pattern recognition, and generalisation during the Exploration and Implementation stages, with debugging during the Verification stage. Two ways in which the students' use of heuristics interacted with CT skills throughout the problem-solving process were observed and characterised as CT -- synonymous and CT -- enhancing.