The Elephant Random Walk (ERW), introduced in the early 2000s, is a fascinating random walk with full memory of its entire past, offering a simple yet rich example of a time-inhomogeneous Markov process. Depending on a memory parameter, its behavior splits into three distinct regimes: diffusive, critical, and superdiffusive. In this talk, we will first introduce the ERW model, then explore its connections to Pòlya urns with random replacement and random recursive trees with Bernoulli bond percolation. Thanks to the urn approach, we will obtain a fixed-point equation for the limiting random variable in the super-diffusive regime, allowing us to extract nice properties of this variable and its behavior. (This is a joint work with Hélène Guérin and Kilian Raschel)