Instruction tuning has been central to the success of recent vision-language models (VLMs), but it remains expensive—requiring large scale datasets, high-quality annotations and large-compute budget. We propose PRioritized cOncept learninG via Relative Error-driven SampleSelection -- {PROGRESS} --a data-efficient framework that enables VLMs to dynamically select what to learn next based on their evolving needs during training. At each stage, the model tracks its learning progress across skills and selects the most informative samples: those it has not already mastered and are not too difficult to learn at the current state of training.This strategy effectively controls skill acquisition and the order in which skills are learned. Unlike prior works, PROGRESS requires no upfront answer annotations, querying answers only on a need basis, avoids reliance on additional supervision from auxiliary VLM, or compute-heavy gradient computations for data selection. Experiments across multiple instruction-tuning datasets of demonstrate that PROGRESS consistently outperforms state-of-the-art baselines with much less data and supervision.