Disclosed are a computer implemented method and system for determining a likelihood of a prospective client to engage in a real estate transaction, by obtaining and/or retrieving one or more characteristics of the prospective client; extracting data regarding a digital interaction behavior of the prospective client; deriving from the retrieved/extracted data one or more digital interaction features of the prospective client directly or indirectly associated with real estate, applying a machine learning algorithm on the derived one or more digital interaction features and on the prospective client's characteristics to determine a probability, a range of probabilities or a category of likelihood of the prospective client to engage in the real-estate transaction.