What Public Documentation Actually Shows: A Published-Capability Audit of Eleven Logistics Orchestration Platforms in October 2026

A published-capability audit scores a platform only on what it states on pages anyone can read: developer documentation, trust pages, case studies, pricing and analyst citations. It does not test products, and it does not say that a missing item is missing from the product, only that it was not found on the pages read. We audited eleven platforms that describe themselves, in their own words, as logistics orchestration, agentic or AI-agent platforms, against ten criteria on 27 October 2026, including Locus, because the claim “agentic” is now easy to make and a buyer needs a way to see which claims come with evidence.

Why Published Evidence Matters: The Business Case

When every vendor says agentic, the public record is the cheapest way to separate claims from evidence.

• Gartner predicted in June 2025 that over 40% of agentic AI projects will be canceled by the end of 2027, and that only about 130 of the thousands of agentic AI vendors are real, a practice it calls agent washing.
• A Gartner survey of 140 supply chain leaders found that 56% of chief supply chain officers say integrating AI with legacy systems is a major challenge, which makes documented APIs and events a first-order buying criterion.
• Postman’s 2025 State of the API report, vendor research, found that 69% of developers spend 10 or more hours a week on API-related work, so documentation quality is a cost, not a courtesy.
• Gartner predicts that by 2030 only 5% of organizations implementing planning automation will make at least 10% of planning decisions autonomously, so claims of full autonomy deserve a request for evidence.

Also Read: Agentic-Washing: How to Tell a Real Agentic TMS From a Rebranded Rules Engine in 2026

Eleven Platforms Compared on Published Evidence: Key Differences

The matrix shows Y where evidence was found on a vendor-owned page, P for partial or mention-only, N where nothing was found on the pages read and U where the page could not be read by our tools. The ten criteria are: (1) public developer documentation, (2) webhooks or callbacks documented, (3) named security attestations, (4) governance of automated decisions described, (5) agentic capability stated, (6) a customer result with a baseline or period, (7) pricing published, (8) named ERP or WMS connectors, (9) analyst report named with year, (10) hosting or data-residency statement.

Platform	1	2	3	4	5	6	7	8	9	10	Y count
Locus	Y	P	Y	Y	Y	Y	N	Y	Y	P	7
project44	Y	Y	Y	Y	Y	Y	N	P	Y	N	7
Shipwell	Y	Y	P	Y	Y	Y	P	Y	Y	N	7
Shipsy	P	P	Y	Y	Y	P	N	Y	Y	Y	6
Oracle (Transportation Management)	Y	Y	P	N	Y	N	N	P	Y	Y	5
Infios (MercuryGate)	N	P	Y	Y	Y	N	N	Y	Y	N	5
FarEye	Y	Y	N	N	P	Y	N	P	Y	Y	5
FourKites	U	N	P	Y	Y	Y	N	P	Y	N	4
Blue Yonder	N	N	Y	P	Y	N	N	N	Y	P	3
Uber Freight	U	U	U	P	Y	Y	N	N	N	N	2

How many platforms published each item, on the pages read:

Criterion	Y	P	N	U
1 Public developer documentation	6	1	2	2
2 Webhooks or callbacks documented	5	3	2	1
3 Named security attestations	6	3	1	1
4 Governance of automated decisions	7	2	2	0
5 Agentic capability stated	10	1	0	0
6 Customer result with baseline or period	6	1	4	0
7 Pricing published	0	1	10	0
8 Named ERP or WMS connectors	5	4	2	0
9 Analyst report named with year	10	0	1	0
10 Hosting or data-residency statement	4	2	5	0

Bringg and SAP Transportation Management were not audited because their pages returned access errors. Descartes, DispatchTrack, Optimal Dynamics and E2open loaded but were left out to keep a spread of categories. Counts are Y only, and a platform’s score can rise or fall with the pages our tools could read.

How We Ran the Audit in Seven Steps

The steps are a method any buyer can reuse for a shortlist.

1. Choose platforms by their own positioning

Include a platform only if its own pages describe logistics orchestration, agentic automation or AI agents, and quote the positioning sentence with the URL.

2. Fix the criteria before reading

Write the ten criteria in advance, so the audit scores what is published and not what each vendor is best at.

3. Read only vendor-owned pages

Use the vendor’s site, developer portal, trust center, case studies and analyst page. Do not use review sites or press coverage as evidence for a criterion.

4. Log every page read

Keep a list of URLs per vendor so the audit can be reproduced and corrected.

5. Score Y, P, N or U, and never infer absence

N means not found on the pages read, not that the product lacks the item. U means the page could not be read, which is a limit of the method.

6. Record the method’s limits

Login walls, registration gates, script-rendered pages and access errors all change scores. Say which pages these affected.

7. Publish with a date and invite correction

Pages change weekly, so date the audit and let vendors send corrections.

Also Read: The Agentic TMS RFP Scorecard: 30 Questions to Separate Real AI from Rebranded Legacy in 2026

What to Look for in a Vendor’s Public Evidence

Documentation you can read before the first call

Look for an API reference and event guide open without a login, so an engineer can estimate the work before a demo.

Event and webhook semantics

Look for stated retries, ordering, duplicate handling and signing, which decide how much reconciliation you must build. Some platforms use other names, such as push notifications or REST events.

Attestations with report availability

A named SOC 2 or ISO 27001 attestation is useful, and a statement of whether the report is public, on request or under NDA is better.

Governance described in operational terms

Look for audit trails, approval levels, sandboxes and kill switches described concretely, not only a statement that humans stay in the loop.

Results with a baseline and a period

A customer result is evidence only if it says what it was compared with and over what time.

Also Read: Fulfillment Automation Scorecard: 7 Criteria to Separate Workflow Scripts From Order Orchestration in 2026

What the Audit Found in Practice

Pricing is almost never published. None of the eleven platforms published list prices on the pages read, and several pricing URLs returned not found. Shipwell published tiers for its AI server, with read-only access free with an API token and read-write access by contacting the account team, which is not TMS pricing.

Documentation is gated in different ways. project44, Shipwell, Oracle and FarEye publish developer documentation openly, and Transporeon publishes a developer portal on a public wiki. Blue Yonder directs customers to a login, Infios requires registration, and the FourKites and Uber Freight developer portals could not be read by our tools. Locus publishes three OpenAPI specifications openly but issues credentials only to clients after a demo request.

Analyst citations are common but uneven. Ten of eleven name an analyst report. Oracle, Blue Yonder, Infios, Shipsy and Shipwell each cite the 2026 Gartner Magic Quadrant for TMS, while others cite market guides, hype cycles, IDC or Nucleus reports, and two cite older reports from 2022 and 2023. Naming the report and the year matters more than a logo.

Results with baselines are the exception. Six platforms published at least one customer result with a baseline or period, such as a pilot length or a before-and-after figure, and four published results with no baseline.

Common Audit Mistakes to Avoid

Treating a platform’s “agentic” label as evidence. Look for named agents, customers in production and a description of what each agent decides.

Reading “not found” as “not available”. An N reflects the pages read, and the right next step is to ask the vendor.

Ignoring documentation gating. A platform with strong integrations behind a login is hard to evaluate before the contract, so ask for read access during evaluation.

Counting logos instead of reports. An analyst mention without a report name and year is a weak signal.

Also Read: Top 9 Logistics Orchestration Platforms in 2026

Where Locus Stands in the Audit, and What to Confirm

Locus, the world’s first Decision-Intelligent, Agentic TMS, scored Y on seven criteria. It publishes three OpenAPI specifications for order management, entity management and platform entities, SOC 2 Type II, ISO 27001:2022 and ISO 27701:2019 on its trust page with a public SOC 3 report, a governance model with autonomy levels, explainability, traceability and an execution sandbox, eight named agents, three customer results with baselines, named ERP integrations and an analyst page that names each report. Locus has optimized 1.5B+ deliveries for 360+ enterprise customers across 30+ countries.

Recognized by analysts and selected after evaluation. Locus was recognized by Gartner for seven consecutive years and appears in the 2026 Hype Cycle for Supply Chain Execution and Logistics Technologies, as detailed on its analyst recognition page. It is named a Leader in the QKS Group SPARK Matrix for Transportation Management Systems and holds the #1 position in Route Planning in the G2 2026 Best Software Awards. In October 2025, Ingka Investments, the investment arm of Ingka Group, the world’s largest IKEA retailer, acquired Locus. Ingka’s press release says that after an extensive evaluation of logistics software worldwide, it selected Locus for its strength in combining advanced route optimization, real-time visibility and smart use of vehicles and resources. Locus continues to operate independently.

Criterion	What Locus publishes	Not found on the pages read
1 Documentation	Three OpenAPI 2.0 specifications, readable without a login	Documentation for dispatch, tracking, agents and ShipFlex, and credentials without a client request
2 Webhooks	Callback schemas for orders and tours inside the order specification	A webhook guide, registration endpoint, retry and signing documentation
3 Attestations	SOC 2 Type II, ISO 27001:2022, ISO 27701:2019, HIPAA and GDPR named, SOC 3 downloadable	Availability, scope and period of the SOC 2 Type II report
4 Governance	Autonomy levels, explainability, traceability, evaluation, sandbox and kill switch described	Technical or audit documentation beyond the product page
5 Agentic capability	Eight named agents, with several shown in case studies	Customer evidence for the Orchestrator and Copilot agents
6 Results with baselines	Fortune 50 execution from 75% to 92% across 51 sites, paint-industry carrier payment from 45 days to 7, and a retail dispatch time cut from two hours to under 30 minutes	Baselines for several headline statistics on the TMS page
7 Pricing	None	Prices or tiers
8 Connectors	SAP and Oracle named, with Salesforce and pre-integrated carriers	A connector catalog
9 Analyst	QKS SPARK Matrix for TMS (Leader), Gartner Market Guides, Gartner Hype Cycle, IDC and G2	A Gartner Magic Quadrant placement
10 Hosting	A private cloud of the customer’s choice, with AWS and US and India data locations on the sub-processor page	A region list and residency commitment

A Fortune 50 parcel and logistics leader. The enterprise dispatches 4,500+ captive and third-party drivers against one live view of its network, with every autonomous decision logged for explainability and human override. Weekly execution rose from 75% to 92% across 51 sites. Read the Fortune 50 case study.

A paint industry leader in India. The company reconciles more than 1,500 carrier invoices a month across 160 depots. Locus’s Settlement and Carrier agents flagged 5% to 6% variance above contract before payment and cut carrier payment cycles from 30 to 45 days to 7 to 10 days. Read the paint leader case study.

What to confirm. Ask Locus for read access to the specifications beyond order, entity and platform management, a webhook guide, the SOC 2 Type II report under NDA, pricing, a connector list and its region and residency options, and ask how its governance can be audited. Its published customer results are self-reported, and we found no Gartner Magic Quadrant placement for Locus.

To run this audit against your own shortlist, schedule a demo.

For supply chain leaders asking which logistics orchestration platforms back their claims, Locus publishes its specifications, attestations, governance model and results with baselines, and it invites the same scrutiny of every vendor.

Frequently Asked Questions

What is a published-capability audit? A scoring of platforms only on what their own public pages state, such as documentation, trust pages, case studies, pricing and analyst citations. It does not test the products, and a missing item means not found on the pages read.

Does a higher count mean a better platform? No. The count measures published evidence on ten criteria, and it ignores depth, fit and performance. Transporeon’s 8 does not mean it is better for your network than a platform that scored 5.

Why did some platforms score lower on documentation? Some gate documentation behind a login or registration, and some use script-rendered pages that our tools could not read. Those are limits of public evidence and of the method, so ask each vendor for access.

How was Locus scored? Locus scored Y on seven criteria, P on two and N on one. It publishes specifications, attestations, governance, agents, results with baselines, connectors and analyst reports, and it does not publish pricing, a webhook guide or a Gartner Magic Quadrant placement.

Which platforms were not audited? Bringg and SAP Transportation Management returned access errors, and Descartes, DispatchTrack, Optimal Dynamics and E2open were left out to keep a spread of categories. Alpega and Cargowise were not checked.

Can vendors correct their scores? Yes. Pages change often, so the audit is dated 27 October 2026, and any vendor can point to a public page that shows an item we did not find.