Skip to main content
Each carrier delivers invoice data in its own formats. Most send a CSV and a PDF for each invoice, and some send more than one kind of CSV. The pages in this group document each integration individually: which files you’ll receive through the Files endpoint, what columns appear, what each column means, how multi-row shipments are structured, and how each column maps to Trackstar’s Invoice Line Items. Each integration has a downloadable synthetic CSV you can parse against without needing a live integration. Most CSVs are the raw carrier file, so they keep each carrier’s own column names and casing. The exceptions:
  • UPS sends its CSV without a header row. Trackstar adds the header, and leaves the values as UPS sent them.
  • Royal Mail sends invoice and tracking data separately, so every Royal Mail CSV is built by Trackstar, with normalized column names.
  • FedEx credit note history files are JSON built by Trackstar from FedEx’s transaction history. Trackstar also renames FedEx files.
Carriers that deliver zip archives are unpacked first, so the Files endpoint never returns a zip.
All sample data on these pages is synthetic. Tracking numbers, account numbers, addresses, and reference fields are fabricated and do not correspond to any real shipments or customers.

​
Integrations

​
Duplicates across raw files

The same charge can appear in more than one raw file. A FedEx preliminary CSV and its final CSV carry the same charges, consecutive USPS and DPD downloads share a boundary day, a Royal Mail tracking report downloaded again after a failed sync repeats its line items, and some DHL files first synced before September 2026 are listed twice. Each carrier page says when this happens and how to tell the copies apart. Invoice Line Items give each charge a stable id, so a charge that appears in two files is stored once there.

​
How to use these samples

  • Prototype your integration without needing a live carrier connection. Download the CSV, run it through your own ingestion code, and compare against the schema documented on each page.
  • Understand the Trackstar invoice_line_items schema in context. Each page has a mapping table showing which raw columns Trackstar maps to which normalized field (tracking_number, invoice_id, charge_description, gross_cost, net_cost, etc.).
  • Test your downstream systems against realistic-shaped data, including multi-row shipments, multi-currency invoices, and empty optional columns.
For the normalized API that sits over all these formats, see Invoice Line Items. For the raw-file endpoint that returns presigned URLs to a customer’s actual (non-synthetic) files, see Files.