01Trade that is still structurally paper-based
And 240 copies for a typical international transaction.
Share of trade documents fully digitised.
Electronic bill of lading uptake in January 2025, against 1.2% in 2021.
With legislation based on UNCITRAL's MLETR model law.
The gap is no longer mainly technological: most paperless trade tools already exist. The difficulty lies in their fragmented deployment and limited interoperability across the transaction chain. Data created digitally in one system must still be re-entered, reformatted or re-checked in another.
Digitisation is not about turning paper into PDFs: it means creating data once, in machine-readable form, and reusing it securely from order to payment. The LC – eBL transaction is a textbook case, as it brings together ordering, transport, customs, bank examination, payment and release of goods.
02A four-layer interoperability architecture
The paper treats the LC – eBL chain as a transactional data system, structured on the UN/CEFACT Buy – Ship – Pay model, and proposes four interdependent layers.
- Legal and regulatoryRecognition of electronic transferable records: the MLETR model law requires identification, integrity, exclusive control and transferability, without prescribing a technology.
- Data and standardsSemantic alignment (what the data mean) and syntactic alignment (how they are structured and exchanged): UN/CEFACT, ICC DSI KTDDE, DCSA, WCO Data Model.
- Identity and trustPersistent identifiers (LEI, vLEI), verifiable credentials, electronic signatures and seals, timestamps: so a document can be relied on outside the system that issued it.
- Connectivity and governanceOpen APIs, multi-party agreements, control-tracking registries: a federated model, with no single platform.
03What already works, and where the architecture is incomplete
| Milestone | What it demonstrates |
|---|---|
| eBL standardisation | DCSA eBL 3.0, ISO 5909:2026 (DLT-based eBL aligned with UN/CEFACT and MLETR) and KTDDE converge: transport-document data are now largely identified. |
| Platform-to-platform eBL transfer (May 2025) | First standards-based transfer between CargoX and EdoxOnline (HMM, Suzano), with no return to paper or bilateral agreement, thanks to the PINT API, the MSPIA agreement and the CTR control registry. |
| End-to-end LC – eBL pilot (2025) | Siemens, Commerzbank, Fr. Meyer's Sohn and a Chinese bank, via trace:original (Enigio) and TradeGo: document processing cut from about two weeks to under two hours. |
These milestones confirm that the Ship phase is the most mature. Gaps remain at the interfaces: Buy-to-Ship, where invoice data are not yet reusable at scale; Ship-to-Border, where eBL and customs data remain weakly connected; Ship-to-Pay, where LC examination still relies on documents rather than data.
The bottleneck is not technical but institutional: the technology existed; the constraint was aligning the parties around shared data standards and mutually recognised legal frameworks.Lesson from the Siemens – Commerzbank pilot
04Towards a global framework: sequencing and enabling conditions
Three sequencing principles: start from “anchor” records (the eBL for Ship, the LC presentation under UCP 600 / eUCP / ISBP 745 for Pay, the structured invoice for Buy); work through cross-document data mapping rather than document by document; aim for federated, technology-neutral convergence.
- Legal reform and accelerated recognitionAccelerate MLETR adoption, including in civil law jurisdictions and major trading economies such as Japan and China.
- Trusted identity, verification and data-exchange infrastructureDefine interoperable trust profiles across trade platforms, customs and banks.
- Awareness, bank adoption and implementation capacitySystems (API-based e-presentation, eBL registry connectivity), training for document checkers, and change management.
- ISO-based alignmentISO 19845 / UBL, ISO 20197-1, ISO 5909, LEI, BIC… and faster adoption of ISO 20022 for trade finance, richer than MT 7xx messages.
- Multilateral governanceCoordinate standards bodies, operators and regulators without a new institution: ICC DSI can act as a coordination platform.
Longer term, integrating programmable payments (wholesale CBDC, tokenised deposits, ISO/TS 23516) would allow a digital equivalent of delivery versus payment, closing the gap between document acceptance and payment. mBridge, Agorá, Ubin+ and Banque de France experiments have demonstrated building blocks, without yet achieving end-to-end integration with the document chain.
Sources : Riham Marii and Julien Achard, Architectures of Interoperability in Digital Trade, policy paper, May 2026, 28 pages (references: UNCITRAL, ICC DSI, DCSA, ISO, WCO, WTO, UNESCAP, UN/CEFACT, BIS).