Tracking Status Codes: 6 Carrier APIs Compared
We counted published tracking status values across 6 carrier and platform APIs as of Nov 2026, with sources, so you can size your status-mapping enum correctly.
Build a unified status field across six tracking APIs and you'll hit the same wall every integration engineer hits eventually: some carriers hand you a five-value enum, others hand you a numeric code table with no stated ceiling. We pulled the current public tracking reference docs for six carrier and platform APIs in November 2026 and counted exactly what each one publishes. The spread is bigger than most mapping layers are built for.
Why carrier tracking API status codes break unified mapping
A normalized status field in your own system has to swallow both ends of this range: a five-value closed enum on one side, an open-ended numeric code table on the other. If you design your mapping switch statement around the narrow case, the wide case overflows it the first week in production. This is the shipment status mapping problem every multi-carrier integration eventually runs into, and it rarely shows up in a sales demo.
Method
We checked each vendor's current public tracking API reference documentation, pulled in November 2026. A "published status" means an enumerated value listed in the vendor's own current reference doc, not a third-party SDK's historical list reused as if it were official. We excluded internal event or scan codes that sit below the top-level status field, deprecated SOAP-era code lists, and aggregator-only normalizations unless we labeled them explicitly as third-party. Where a vendor doesn't publish a single finite enum, we wrote "Not published" rather than estimate one from sample responses.
The table: six APIs, six different taxonomies
| API / Platform | Status model type | Published value count | Example values | Source |
|---|---|---|---|---|
| DHL Shipment Tracking – Unified API | Fixed enum | 5 | pre-transit, transit, delivered, failure, unknown | DHL API Support |
| FedEx Supply Chain GetTrack | Fixed enum | 6 | OC, PU, AR, DP, OD, DL | FedEx Supply Chain Developer Production |
| DPD API (v1.2.5, March 2025) | Numeric internal status codes + separate additional codes | Not published as a single finite list | 02 In terminal; 13 Delivered to consignee | DPD API documentation v1.2.5 PDF |
| UPS Tracking API (REST) | Raw statusCode / type fields, no documented closed enum | Not published | code: SR, type: X, simplifiedTextDescription: Delivered | UPS Developer Portal / UPS-API GitHub reference |
| PostNord, via AfterShip (not PostNord's own docs) | Third-party aggregator tag enum | 9 | Pending, InfoReceived, InTransit, OutForDelivery, AttemptFail, Delivered, AvailableForPickup, Exception, Expired | AfterShip PostNord Logistics API docs |
| EasyPost Tracker / Shipment status | Fixed enum (platform-normalized) | 10 | unknown, pre_transit, in_transit, out_for_delivery, delivered, available_for_pickup, return_to_sender, failure, cancelled, error | EasyPost Docs, Tracker object |
Checked date for all rows: 17 November 2026.
DHL and FedEx: tight, documented, low cardinality
The status codes for the Shipment Tracking - Unified API are the following: pre-transit, transit, delivered, failure, and unknown. DHL's own support article even documents what each one means at the definition level, down to why a failure event fires. FedEx Supply Chain's GetTrack endpoint runs a similarly tight list: OC for shipment information sent to FedEx, PU for picked up, AR for arrived at FedEx location, DP for departed FedEx location, OD for on FedEx vehicle for delivery, and DL for delivered. Six codes, one field, no ambiguity about where the enum ends.
DPD and UPS: open-ended by design
DPD doesn't publish anything you could call a finite enum. StatusCode is used for internal DPD processes to identify parcel life cycle status, and the doc walks through dozens of numeric codes individually, e.g. "02 In terminal Parcel was accepted in terminal" and "13 Parcel is delivered to consignee". Worse for mapping purposes: the same terminal delivery code is reused for both successful delivery and return-to-sender, so you can't read status in isolation. The doc is explicit that combinations of statusCode and serviceCode are needed to understand the correct status of the parcel.
UPS sits in a comparable spot. Its current Track API reference returns a currentStatus object with a short human description, a numeric statusCode, and a type flag, for example "description": "Your package was released by the customs agency.", "simplifiedTextDescription": "Delivered", "statusCode": "003", "type": "X". There's no page in UPS's own docs listing every valid combination as a closed set the way DHL does. Treat any "UPS has 40 status codes" claim you find online as a third-party SDK's historical sample, not a documented contract.
What "Not published" actually costs you
For UPS and DPD, "Not published" isn't a documentation gap you can just shrug off. It means:
- You cannot write an exhaustive switch statement against either API and trust it stays exhaustive. New codes can appear in production that your mapping layer has never seen.
- For DPD specifically, you need at least two fields (statusCode plus serviceCode) to resolve one logical status, not one.
- For UPS, teams generally fall back to the
typeflag alongsidestatusCodeand the human-readablesimplifiedTextDescription, since none of the three alone is documented as a stable closed set. - Any third-party list claiming to be "the UPS status codes" or "the DPD status codes" should be treated as a snapshot, not a spec.
Build the mapping for the widest case, not the narrowest
If you're integrating DHL's five-value enum and DPD's open-ended numeric system behind one internal status field, your mapping layer needs a catch-all or overflow bucket, not just a clean 1:1 switch. A new DPD code that doesn't match anything you've mapped shouldn't silently default to "unknown" and disappear from your dashboards, it should land somewhere visible. The same principle that applies to idempotency keys and webhook retries applies here: store the carrier's raw code alongside your normalized bucket, always. You will need it the first time a customer asks why their DPD shipment shows "in transit" when DPD's own tracking page says something else entirely.
Where multi-carrier platforms absorb this for you
Several platforms exist specifically to do this normalization so you don't maintain six separate mapping tables yourself: ShipEngine, EasyPost, Shippo, Sendcloud, nShift, AfterShip, ClickPost, and Cargoson all ship their own internal status taxonomy that sits between the carrier's raw codes and your application. EasyPost's own ten-value enum, shown in the table above, is a direct example of what that normalization looks like once it's built: unknown, pre_transit, in_transit, out_for_delivery, delivered, available_for_pickup, return_to_sender, failure, cancelled, error. That doesn't mean every platform's normalization is equally complete or equally accurate against what the carrier actually sends. We haven't benchmarked coverage or mapping accuracy across these platforms yet, that's a separate test and one we'd want real webhook samples for before publishing a verdict.
Caveats and the obvious follow-up
Documentation changes without changelog entries on more than one of these vendor portals, so re-check quarterly if this table is feeding a production mapping decision. We also didn't test runtime behavior here, only documented surface area. Whether DHL's API in practice ever emits anything outside its five documented values, or whether UPS's statusCode space is actually bounded in observed traffic even without a published list, is a real benchmark worth running next: sample a few thousand live webhook payloads per carrier and diff them against the documented enum. If you've already run that test on your own traffic, we'd like to compare notes.