Data warehouses must homogenise and integrate data from an organisation's areas to extract relevant knowledge for orientating decision-making. This is not an easy task, which is why several approaches have been developed which can be classified according to how information is obtained: supply-driven, demand and a hybrid of the first two. This study offers a conceptual framework and a chronological study of the approaches adopted according to the paradigm used, providing a comprehensive context for better understanding the current state of data warehouses and their evolution.