If you have a look at the data that is stored for you on the SFTP-server you will find the following folder structure. If you navigate to a specific day you will find two files: <filename>_json.zip and <filename>_csv.zip.
If you then decompress one of these archives you will find - separated to individual files - the data that was tracked that day. If you have a lot of traffic than the files will be split into multiple smaller files instead of one big one → example: user1.json, user2.json, user3.json.
The JSON-archive includes the complete raw data of that specific day stored as .json files. Most JENTIS users use the .csv files to get started quickly in a flat file format and store the .json files into their own data lake to have it ready for future use cases.
The CSV-archive includes .csv files with a sub selection of the most frequently used variables. These .csv files are convenient to be used for most general use cases as the nested properties of the JSON are already spread into separate columns.
The following objects and fields are provided in the raw data. Each document (objects that were submitted via _jts.push track command) gets its own file when decompressed. All JENTIS Raw Data outputs have at minimum the documents: user, session and event. Additionally there are all the documents that are optional (pageview, product, order, etc).
A document in the JENTIS internal data model is an object that is tracked. The system internally generates two mandatory documents on all activities: user and session. Additionally each object that is pushed to the JENTIS Data Layer is considered a document, ie. "pageview", "event", "product" or any document you define in your tracking implementation process. This is based on the "track" command that is described in detail here: JENTIS Data Layer Fundamentals.
Each document has entries with the following fields.
Per default the user document doesn’t have any additional fields.