The World Wide Web is today's richest medium of information and services. Interchanging data freely without any time constraints is much more than appealing. However, the Web has to be restructured in order to address today's needs and expectations, favouring users' searches and navigation and, to do so, Web analysis tools are the order of the day. There is a similarity between the analytical Web data processing and the analytical processing of conventional data. The truth is that the enormous volume of data along with the high transactional activity strongly prohibits the use of raw data. In such a scenario, Web server logs emerge as the main sources of information and there are many Web analysis tools today that work over these raw elements. However, the obtained results are far from satisfactory, since they do not help to understand users' navigational patterns. In the near future, data webhouses will be in charge of any analytical work, requiring well-defined extraction, transforming and loading processes, capable of reuniting all relevant data and ensuring its consistency and quality. Only when these processes are implemented is there room for thinking about Web Mining. This paper aims to target the problem of establishing data webhouses as primary Web Mining sources, highlighting the excellence of the Web domain towards mining and defining the guidelines that populating processes must comply with. In this sense, a case study will be presented, identifying its available data sources, establishing a proper data webhouse and specifying all the processing required to populate it.