It proves that the MLE is the most powerful estimator.

It defines the most powerful test as having a rejection region based on the likelihood ratio $\frac{L(\theta_0)}{L(\theta_1)} < k$.

It guarantees that the test is UMP for composite hypotheses.

It provides a method to calculate the p-value without knowing the alternative distribution.