Historical Data

Rockerbox distinguishs between new and repeat customers, allowing you to understand the unique marketing mix and path to conversion for your new vs. repeat users.

In order to do so, Rockerbox needs a one-time log of all historical orders.

If you leverage Shopify, Rockerbox automatically pulls in your historical orders, and you do not need to separately send us a file.

The dataset we require includes the following:




Match the format from the Purchase event (i.e. don’t send us Shopify Order IDs from your Purchase event, and CRM Order IDs for the historical file).


Match the format from the Purchase event.


Optional. We can use this value to match users in cases where user_id formats change over time (ex: you switched from one order system to another).


Full timestamp of order including date, time, timezone offset—ideally in UTC.


This should match the revenue value we get on the Purchase event. If you send us net revenue, the historical file should also be net revenue. if you send us gross, it should be gross.

  • File Format. The file will be in a spreadsheet format, such as .csv, .xlsx, or gsheet. If the file is too large to send over email, Rockerbox will setup a shared folder for the customer to drop the file.
  • Date Range. The date range of the historical order file should start at the beginning of the customer’s order history all the way to when the Purchase event was completely setup in Rockerbox.
  • File Size. If your order history contains over 1 million records, break your files up into files containing 1 million records each. Keep the column headers consistent across the files.
  • Processing Time. It will take Rockerbox approximately 3 business days to process the file, assuming we do not encounter any issues in the dataset.
  • User Identifiers. Rockerbox only requires one of user_id or email_address to process the historical order file but recommends including both fields in the file.

How did we do?