Web design has been an important part of websites for almost two decades now. It gives identity to websites and tends to deliver information to end users in a structured, organized and elegant way. Occasionally it also fails, delivering poor experience to end users who want to access information they are looking for. In this thesis, we model a relationship between the attributional description and user experience of web design, focusing on the aesthetic point of view. Our goals are the following: to define attributes which can describe web page design in a suitable way, to collect a set of websites, calculate their descriptors, supplement them with human-based aesthetics ratings through crowd-sourcing, and model the relationship between the design and the ratings by using machine learning models. We use the best performing model to develop a Chrome extension prototype that will give users the possibility to evaluate and analyze the design of websites.