# Usage data ## General Logs are kept for the following types of usage, for invoicing purposes: * [**DeepL**](https://www.deepl.com) **translation requests**: user ID, timestamp and number of characters. The actual query and the answer returned by DeepL is not saved. * **LLM requests**, either with ClauseBase's key or with the customer's key: user ID, timestamp and number of tokens. * **OCR requests** (conversion of scanned PDF documents): user ID, timestamp, outcome (result or failure), error messages and number of pages. * **Exported documents** (DOCX or PDF): user ID, template number, template title, timestamp, number of pages, whether the end-user interface or instead the API was used, optional identifier given by the user and ID of the brower session. The content of the exported documents is not saved. ## ClauseBuddy On the basis of the data stored in the [central logging system](https://legal.clausebase.com/data-collected/logging-data) each customer's administrator can request an overview of the usage of ClauseBuddy by each of its users. The administrator can get an overview of the following data: * Amount of clauses, templates, users and saved answers being created, updated and stored at a given moment. * Which clauses were most frequently inserted into MS Word documents by users. * How many times ClauseBuddy was opened by each user. * How many keyword-based searches and browse sessions were performed in ClauseBuddy's Quality Library. * How many AI-related requests were made.

In addition, administrators can check the time patterns of its users — i.e. at which times of the day that ClauseBuddy was used in the company, on a per-user basis. For privacy reasons, this usage is grouped per block of 15 minutes.