There are several ways to prevent web data extraction, which can be combined and operate differently. One such approach involves declarative methods, meaning they do not block a bot’s access to the website but instead specify access rights in a non-binding way. The simplest method is to include provisions in the website’s terms of use (CGUs). These allow for a general objection to data collection the website, but they are costly to read. Indeed, this would require collecting the terms of use for each scraped domain, parsing their content, and extracting the rules governing access to the rest of the website. Such a practice is error-prone due to its lack of standardisation and does not scale well. It therefore needs to be complemented by machine-readable methods.