Average BLS runtime as a function of the time sampling rate of input data. Three algorithms are compared: a brute-force AstroPy BLS (red), FFA-based GERBLS (green), and GERBLS with automatic data downsampling (blue). GERBLS runtimes have a near-quadratic inverse dependence on the time sampling rate: a shorter cadence leads to an increase in the number of tested periods and transit phases. As demonstrated in Section 3.2, the sampling time can usually be increased as long as it remains shorter than the duration of any real transit signals in the data. In all cases, GERBLS significantly outperforms AstroPy in average runtime.