rainforest canopy layer top view
← Back to blog

Why Blockchain Failed at Supply-Chain Traceability (and What Actually Replaced It)

Company News•27 October 2026•TREEO Indonesia•5 min read

Between 2022 and 2023, at least six blockchain-based supply-chain and trade-finance platforms — each backed by major institutions, some running for years, one with over 175 participating organizations — shut down. Not scaled back. Shut down. The pattern is consistent enough, and well-documented enough, that it's worth understanding exactly what failed and why, especially now that EUDR and similar compliance regimes are pushing a new wave of interest in "put it on the blockchain" as a traceability answer.

What actually happened

6

named blockchain trade/traceability platforms that failed, 2022-2023

175+

organizations that had joined the largest one (TradeLens) before it shut down

$0.23

average Ethereum transaction fee today — the "too expensive" argument no longer applies

0

production zero-knowledge-proof traceability deployments with real users, as of today

The Casualty List

TradeLens (Maersk + IBM, launched 2018 on Hyperledger Fabric) was the largest attempt: at its peak, 175-plus organizations, 600-plus ports and terminals, five of the six largest ocean carriers. It was discontinued in November 2022, with IBM ending access in March 2023.

we.trade (12 European banks plus IBM, live from 2019) posted an $8 million loss on roughly $4 million in revenue in 2020, with only 2 of its 12 shareholder banks fully deployed. It went into insolvency in June 2022.

Marco Polo Network (TradeIX/R3 Corda, 30-plus banks) missed its 2019 and Q2 2020 go-live targets and entered provisional liquidation in Ireland in February 2023, with €5.2 million in debts.

Contour (letters of credit, which had bought we.trade's rulebook after its collapse) ceased operations in November 2023 after failing to raise further funding.

B3i (an insurance-industry consortium founded in 2016 by Aegon, Allianz, Swiss Re, Munich Re, and Zurich) shut down in July 2022, citing insufficient volume and commitment.

Agrikore (Cellulant, built on Hyperledger for agricultural payments in Nigeria) discontinued its marketplace in November 2020, following an internal fraud review. The smart contracts reportedly worked correctly; the human process controls around them — KYC, treasury management, segregation of duties — did not.

PlatformSectorOutcome
TradeLensContainer shippingDiscontinued Nov 2022
we.tradeTrade financeInsolvent June 2022
Marco Polo NetworkTrade financeLiquidation Feb 2023
ContourTrade financeCeased operations Nov 2023
B3iInsuranceShut down July 2022
AgrikoreAgricultural paymentsDiscontinued Nov 2020

One consortium platform in the same category, komgo (commodity trade finance, founded 2018 with BNP Paribas, Citi, Shell, BP and Glencore among its backers), is still running — but survived by de-emphasizing blockchain entirely and repositioning as a conventional "digital network" workflow product.

What the Failures Have in Common

Across independently reported accounts of each collapse, the same causes recur. Consortium governance owned by one or two dominant participants repels the competitors it most needs to join — a platform co-owned by a shipping line struggles to convince rival shipping lines to feed it data. Funding runs out in the gap between an impressive pilot and paying-customer volume; we.trade's own board minutes described roughly 18 months of runway on its final funding round before the shutdown. And in every documented case, the network never crossed the threshold where the value of joining exceeded the cost of joining — the fundamental network-effect problem, unsolved by better cryptography.

The pattern that survived — komgo's pivot — is instructive on its own: keep the workflow value, drop the "blockchain-first" framing that was never actually the point.

The Deeper Problem: A Ledger Proves What Was Written, Not What Was True

Beyond the specific governance failures, there's a structural limit worth naming plainly, because it applies regardless of how well a future platform is run. A blockchain — permissioned or public — cryptographically guarantees that a recorded entry hasn't been altered after the fact. It guarantees nothing about whether that entry was accurate when it was written. If a shipment's origin data is wrong at the point of entry, an immutable ledger simply immortalizes the wrong data instead of allowing it to be corrected. This is sometimes called the "oracle problem" in the peer-reviewed sustainability-metrics literature: no mechanism yet exists that cryptographically binds a real-world physical event to a ledger entry without a trusted human or institution vouching for it at the point of entry. That's a category limit, not an engineering gap waiting to be closed.

There's a second, more concrete problem specific to compliance contexts like EUDR: geolocation data linked to identifiable smallholder farmers is personal data under GDPR. Writing immutable, identifiable plot coordinates to a permanent ledger collides directly with data subjects' rights to erasure and rectification — rights that a genuinely immutable record structurally cannot honor. The EU's own implementing regulation for EUDR's Information System caps personal-data retention in due-diligence records at five years, which only makes sense if that data is expected to eventually be deletable — the opposite of what a blockchain guarantees.

What Actually Works: Standards, Not Ledgers

The useful finding buried under all of this is that the underlying problem — proving where something came from, in a way a third party can verify — is already solved, using ratified, boring, non-blockchain standards that mostly predate the current AI/blockchain hype cycle entirely.

The W3C's Verifiable Credentials Data Model 2.0 became a full Recommendation in May 2025: an issuer can cryptographically sign a claim (a certificate, an origin attestation, a due-diligence reference number) in a way that's tamper-evident and independently verifiable — with no ledger required at all. GS1's EPCIS 2.0 standard, ratified in 2022 and adopted as ISO/IEC 19987:2024, is the deployed, production-grade format for sharing "what, when, where, why, and how" event data across supply-chain parties — and GS1's own guidance explicitly recommends anchoring only a hash to any blockchain, never the actual event data, precisely because of the problems above.

Selective disclosure has its own solved layer: SD-JWT (IETF RFC 9901, published November 2025) lets a credential holder reveal only the specific claims a verifier needs, rather than an entire document — already mandated for the EU's own digital identity wallet framework. A more advanced form, BBS+ signatures, would add full unlinkability between separate disclosures of the same credential, but remains blocked on a pending cryptography standards finalization as of this writing — worth watching, not yet ready to build on.

Notably, the sectors that have actually delivered measurable traceability progress ahead of regulatory deadlines did it without blockchain at all. Cocoa traceability initiatives report 82-83% of direct cocoa traceable to farm level in Ghana and Côte d'Ivoire as of 2024, and individual companies have polygon-mapped hundreds of thousands of farms — all using conventional GIS, databases, and satellite verification. Indonesia's timber-legality system (SVLK) and the FLEGT licensing scheme it underpins have run as accredited, auditable government systems since 2016, with no blockchain component. Where traceability has actually been delivered at scale, the tooling has consistently been polygons, databases, and satellite checks — not distributed ledgers.

Frequently Asked Questions

The major named trade-finance and shipping-traceability consortiums did — TradeLens, we.trade, Marco Polo Network, Contour, and B3i all shut down between 2020 and 2023. One (komgo) survived by dropping the blockchain-first framing and repositioning as a conventional digital workflow platform.

Provenance without the ledger

The pattern that actually works — signed attestations, structured event data, and reproducible pipelines — is exactly what defensible evidence for compliance work requires.

Blockchain Traceability: Why It Failed, What Works Now | TREEO Indonesia