Degradation of dye solar cell performance based on the early changes in electrolyte color is predicted, allowing to estimate the lifetime of the dye solar cells even before their efficiency declines. Previous predictive models commonly rely on regression analysis of the predicted parameter; thus, they are unable to capture degradation before a significant decrease in performance. Degradation tests, even when accelerated, may take thousands of hours. As such, recognizing degradation trends early can lead to rewarding cuts in the duration of solar cell development pipelines. With accurate lifetime predictions, researchers can steer materials research to reach longer lifetimes in shorter cycles. The predictive power of our model relies on color changes in the electrolyte that directly correlate with the concentration of tri-iodide charge carriers within it, the loss of which is the predominant degradation mechanism for most liquid-electrolyte dye solar cells. By linking the physical mechanisms inside the cell, which eventually start to degrade the performance of dye solar cells, an early prediction of the lifetime can be made even when the device performance still appears stable. It is exemplified with dye solar cells that integrating architecture-specific knowledge on degradation mechanisms has potential to improve lifetime predictions for photovoltaics.