This dataset contains the processed and aggregated departmental-level data used in the productive scale segmentation analysis of Colombian transitory crops for the period 2007–2024. It is derived from the Evaluaciones Agropecuarias Municipales (EVA), published by the Colombian Ministry of Agriculture and Rural Development (MADR). The dataset includes 15,077 records corresponding to 490 department–crop pairs across 56 crops and 32 departments, representing 90.65% of total cumulative production in the sector. Each record contains semi-annual observations of planted area (ha), harvested area (ha), production (t), and yield (t/ha), together with engineered features used for K-Means clustering: average production, average planted area, number of active periods, temporal Herfindahl–Hirschman Index (HHI), and spatial HHI. The clustering results assign each pair to one of three productive scales (large, medium, or small). A supplementary file includes the full segmentation output with cluster labels and descriptive statistics per group.