Locus Argues for an Orchestration Layer Above Carrier APIs

Locus Argues for an Orchestration Layer Above Carrier APIs

Locus published a piece on API integrations for logistics platforms built around a number that should be familiar to anyone who has run a multi-carrier operation: enterprises accumulate 15 to 30 point-to-point integrations — FedEx, UPS, DHL, USPS, plus the WMS, the OMS, tracking, and settlement — and then spend their engineering capacity absorbing each carrier’s version changes. The argument is that the integrations themselves are not the advantage. Every competitor has them. The advantage is an orchestration layer above them that consumes the standardized data and makes allocation and routing decisions in real time, rather than passing information downstream to a planner who is already too late to use it.

The post’s supporting figures are vendor figures and should be read as such: $288 million recovered by Locus enterprises through carrier settlement matching, a 72% gain in carrier allocation efficiency, and a market moving from $6.2 billion in 2025 to a forecast $16.9 billion by 2034. The claim that survives independent of the marketing is the architectural one, and it has a specific tell — the post says the optimizer solves across 180+ variables simultaneously. That number also appears in Locus’s catalog profile as the stated scope of its dispatch planning and route optimization, which is a rare case of a marketing figure and a catalog record agreeing on the same claim.

What is interesting is that Locus’s own API surface is shaped exactly like the layer the article describes. The catalog holds sixteen API pages, and almost all of them are master-data resources rather than transactions: LocationMaster, PersonnelMaster, SKUMaster, TeamMaster, Vehicle, ServiceType. The transactional pair — Order and OrderServiceability — sits on top, with Transporter handling the carrier abstraction the post names as the 3PL priority. That is the constraint model an optimizer needs, exposed as an API. The orchestration layer is not a metaphor here; it is the product’s actual shape.

Locus scores 36.0, thin, on the Kin Score — good discoverability (74.1) and real contract quality (49.0) undercut by governance at 0.0 and operational transparency at 2.6. No public changelog, no status page, no versioning policy on a platform whose entire pitch is that you should route your carrier decisions through it. The agentic access profile reads 48 operations, 34 acting and 10 human-in-the-loop, which is a fair split for a system that dispatches vehicles. The article’s own closing advice — that security and observability matter as much as endpoint coverage — is the right advice, and it points directly at the two facets where Locus’s score is empty.

← Banuba Moves the KYC Reject Loop Onto the Device
WorkOS: A Transient JWKS Failure Is Not a 401 →