TradeLens is gone, and the way it ended tells you more than the way it launched. When Maersk and IBM said they would wind the platform down, a lot of the coverage filed it under "blockchain hype dies." I read it differently. I have spent years arranging cross-border loads for small and mid-sized shippers, and to me TradeLens was never really a technology story. It was a governance story wearing a technology costume. The ledger worked well enough. The consortium behind it did not hold together.
This piece is a retrospective, not a eulogy. I want to walk through what the platform set out to do, why two of the most capable companies in shipping could not make it stick, and what those of us who still have containers to move next week should actually take from the whole episode. Maersk announced the decision to discontinue TradeLens on 29 November 2022. The platform was set to go offline by the end of the first quarter of 2023, and IBM ended access to it as of 31 March 2023.
If you were a shipper or forwarder who had wired shipment events into that pipe, the shutdown was a scramble. If you were watching from the sidelines, it was a very expensive lesson in a truth that logistics people already know in their bones: a network is only as useful as the people willing to join it. TradeLens built the road. It could not persuade enough drivers to use it.
What the end of TradeLens actually taught us about digitising trade
Here is the short version before the long one. TradeLens was a permissioned data-sharing network for container shipping, jointly created by IBM and GTD Solution, a Maersk subsidiary. It launched under the TradeLens name in 2018, built on the IBM Blockchain Platform, which runs on the open-source Hyperledger Fabric ledger. Its pitch was clean: put every party in a shipment (carrier, port, terminal, customs authority, forwarder, bank) on one shared, tamper-evident record so nobody has to email PDFs back and forth or reconcile three versions of the same bill of lading.
For a while the pitch worked. By the end of 2019, Maersk reported that five of the six largest ocean carriers planned to be on the platform, after CMA CGM and MSC joined in May of that year and Hapag-Lloyd and Ocean Network Express followed in July. Between them, the participating carriers touched more than half of the world's ocean container cargo. On paper, that is dominance. In practice, it was the high-water mark, and the tide went out from there.
What I take from the arc is uncomfortable for anyone selling a "single source of truth." The technology was not the bottleneck. The barrier was trust between competitors, the economics of onboarding, and the plain difficulty of getting rival firms to route their most sensitive operational data through a system that carried a competitor's name on the door. Those are human and commercial problems. No consensus algorithm solves them.
So this retrospective is organised around the questions a practitioner asks, not the ones a vendor asks. What was it for. Why did it die. Who felt the pain. And what would I do differently if someone handed me a blank sheet and a budget today.
What TradeLens set out to build
The core promise was shared visibility. In container logistics, the same shipment generates dozens of documents and status updates as it passes from shipper to trucker to terminal to vessel to customs and back to a warehouse. Each handoff is a place where data gets re-keyed, delayed, or lost. TradeLens wanted to replace that relay race with a single permissioned ledger where each event was written once and every authorised party could read it in near real time.
Three use cases carried the weight of the vision. End-to-end shipment visibility, so a shipper could see a container's status without phoning three intermediaries. Digitised trade documents, with the electronic bill of lading as the marquee example. And faster customs handling, where pre-arrival data could let authorities clear cargo before it hit the quay rather than after.
The privacy design was more thoughtful than critics later gave it credit for. Maersk was explicit that a carrier's data was not visible to its competitors. Ocean carriers could run their own blockchain nodes for the consignments they controlled, and access permissions let each data owner decide exactly who could see what, even after a record joined the shared ledger. On the architecture, this was sound. Owners kept control of their data. The ledger just made the sharing auditable and hard to tamper with.
And yet the thing that made TradeLens technically credible is exactly where the commercial trouble hid. A permissioned network needs permission, which means it needs willing members. The engineering assumed cooperation as an input. The market treated cooperation as the hardest thing to obtain.
Why Maersk and IBM pulled the plug
The official reason is on the record and worth quoting plainly. Rotem Hershko, then Maersk's head of business platforms, said that while TradeLens had successfully developed a viable platform, the need for full global industry collaboration had not been achieved, and as a result the platform had not reached the level of commercial viability necessary to continue as an independent business, according to Maersk. Read that twice. The company is not saying the software failed. It is saying not enough of the industry showed up.
Why didn't they show up? This is where my experience in the trenches lines up with what analysts concluded. The governance model made Maersk a co-owner of a network that its direct rivals were being asked to feed with operational data. Even with airtight technical privacy controls, a competing carrier has to weigh the optics and the strategic risk of routing its shipment flows through infrastructure branded by the largest player in its own market. Neutrality was the missing ingredient, and neutrality is a governance choice, not a coding task. Commentators including Forbes made this point at the time of the shutdown.
Then there is the return-on-investment problem, which gets less airtime but matters just as much. Getting a carrier, a port, or a forwarder onto a network like this is not a plug-and-play afternoon. It means mapping legacy systems, building integrations into terminal operating systems and customs interfaces, cleaning up data that was never standardised, and training staff. Those costs are real and they land up front. The savings are diffuse and they arrive later, if participation holds. When too many partners hesitate, the shared benefit that justifies the shared cost never fully materialises, and the business case quietly collapses under its own onboarding overhead.
Scale compounds the same issue. A network's value grows with membership, but so does the coordination burden. Every new participant is another set of data formats to reconcile, another legal review of data-sharing terms, another node to keep in sync. TradeLens proved it could sign up marquee names. It could not convert enough of the long tail (the mid-tier carriers, the regional forwarders, the smaller ports) into daily contributors fast enough to reach self-sustaining economics before patience and funding ran out.
None of this makes the project a folly. It makes it an experiment that returned a clear, if expensive, result. When the people running it concluded the numbers would not work as a standalone venture, they shut it down rather than let it limp. I respect that more than the alternative.
What the shutdown meant for shippers, carriers, forwarders, and customs
When a shared platform disappears, the visibility it provided does not transfer to a successor by itself. It fragments back into the bilateral relationships it was meant to replace. The window between the announcement and the shutoff is where the real risk sits, because that is when data flows go dark before replacements are wired in. When we arrange cross-border loads for smaller shippers, that transition gap is precisely where exceptions turn into detention charges and missed sailings. Acting early on direct data links beats waiting for the industry to bless a single replacement.
For shippers, the practical move was to rebuild visibility through direct API connections with carriers and forwarders, so that shipment milestones kept flowing into their own systems. The prize is faster exception handling, which is the lever that actually reduces detention and demurrage. Anyone moving temperature-sensitive cargo had a sharper version of the same task: make sure cold-chain readings and chain-of-custody events still reach downstream systems without the shared ledger as the middleman.
For carriers, the shutdown meant accelerating direct integrations and settling on common data models with trading partners. The friction that TradeLens had absorbed (disputes over documents, mismatched fee structures, arguments about which event happened when) came back onto bilateral rails. Carriers that had leaned on the platform for a clean shared record suddenly had to reconstruct that reconciliation partner by partner.
For freight forwarders, the coordination load is heaviest, because a forwarder sits between many carriers and many shippers at once. The answer was standardised event streams and as much automation as possible in load matching and document reconciliation, plus a contingency plan for the moment a partner restricts data access. Forwarders who had treated TradeLens as their integration backbone had the most rewiring to do.
For customs authorities and regulators, the lesson pointed toward standards rather than platforms. A minimal, openly published data-exchange standard that any party can implement is more durable than a single proprietary network, however well built, because it does not depend on one venture staying solvent. The cleaner path is to require machine-readable filings against an open specification and let vendors compete to deliver them.
Across all four groups the underlying instruction was the same. Do not make your visibility depend on any one platform surviving. Build it on interfaces you control and standards you can switch vendors around.
The lessons I took from the TradeLens experiment
Settle the governance before you write a line of code. This is the lesson I keep coming back to. TradeLens got the sequence backwards for the market it was trying to serve. The technology was built, and then the venture tried to convince competitors to trust the ownership structure. In my experience coordinating carriers across corridors, the order that works is the reverse: agree who owns the network, who arbitrates disputes, and who is barred from seeing what, and only then build. If the answer to "would a rival route its data through this?" is "probably not," no amount of cryptography rescues the project.
The second lesson is about neutrality specifically. A shared industry utility tends to need a neutral steward, whether that is a genuinely independent consortium, an industry body, or a structure where no single competitor holds the pen. The digital bill of lading standards work advanced by neutral groups such as the Digital Container Shipping Association has arguably had more staying power than TradeLens precisely because it sells a standard everyone can adopt rather than a platform one firm owns.
The third lesson concerns where the money and the pain sit. Digitisation projects routinely underestimate change management relative to the technology. Getting data clean, getting staff to trust the new flow, and getting legacy systems to talk is slow, unglamorous work, and it dwarfs the cost of the ledger itself. The early trade-digitisation pilots that stalled, TradeLens among them, tended to stall on adoption and integration, not on the code. FreightWaves tracked the carrier sign-ups closely at the time, and the pattern in that reporting was that momentum lived and died with who agreed to participate.
The fourth lesson is more hopeful. Interoperability beats consolidation. The World Bank Logistics Performance Index has repeatedly flagged data connectivity as a differentiator for how efficiently a trade corridor performs. That is an argument for open standards that let many systems exchange data cleanly, not for one platform to swallow the field. A world with several interoperable networks that speak a common language is more resilient than a world betting on a single winner, because when one venture folds, the standard survives it.
If I were scoping a pilot today, I would keep it modular and narrow. Pick a corridor, agree the field definitions for a handful of events and documents, publish the governance charter first, and measure a small set of outcomes that a CFO recognises: transit time variance, dispute rates, and how fast exceptions get resolved. Prove value on one lane before asking anyone to bet a network on it.
Where digitisation goes next: standards over platforms
The direction of travel is clear enough: shared standards, not shared owners. The most useful thing to require of every partner is machine-readable records for cargo, condition data, and shipment events against an open specification that any system can implement. That baseline lets data move cleanly between systems and, crucially, lets you swap vendors without ripping out your visibility.
A practical data model for this covers the things that actually cause reconciliation headaches: shipment identifiers, container and seal numbers, processing and customs status, and condition readings for sensitive cargo. Consistent identifiers for shipments, parties, and locations do more to reduce manual reconciliation than any ledger technology, because most disputes trace back to two systems calling the same thing by two different names.
On the plumbing, the goal is interoperable interfaces with common semantics: APIs and standard event formats that connect transport, warehouse, and enterprise systems and reach across brokers, forwarders, and carriers. Blockchain still has a legitimate role where a tamper-evident, multi-party audit trail genuinely adds value, but it should sit inside an open-data approach rather than being the reason the project exists. The failure mode to avoid is leading with the technology and hoping the participants follow.
Governance deserves the same seriousness as the architecture. Clear roles, published rules for data quality and access, defined audit trails, and an escalation path for when something breaks. Endorsement from industry bodies and regulators helps, because it lends the neutrality that a single-company platform struggles to claim. This is the part TradeLens got wrong for its market, and it is the part any successor has to get right.
At GetTransport.com we live in the messy present this all points to, arranging moves across borders where visibility still depends on stitching together carrier feeds, forwarder updates, and customs status by hand. That daily reality is the honest test of any grand digitisation plan. A standard that survives a vendor going bankrupt is worth more to a working shipper than a beautiful platform that might not be here next year. TradeLens taught the whole industry that lesson at considerable expense, and the least we can do is learn it.
Was TradeLens a blockchain, and who owned it?
Yes. TradeLens ran on the IBM Blockchain Platform, which is built on the open-source Hyperledger Fabric, a permissioned blockchain. It was a joint effort between IBM and GTD Solution, a Maersk subsidiary, and it launched under the TradeLens name in 2018. Being permissioned meant participants were known and vetted rather than anonymous, and data owners controlled who could read their records.
When exactly did TradeLens shut down?
Maersk and IBM announced the discontinuation on 29 November 2022, according to Maersk. The platform was scheduled to go offline by the end of the first quarter of 2023, and IBM ended access to it as of 31 March 2023.
Why did TradeLens fail if major carriers had joined?
Adoption peaked but never became universal. Maersk's stated reason was that full global industry collaboration was not achieved, so the platform did not reach commercial viability as a standalone business. The deeper issue widely cited by analysts was neutrality: competing carriers were reluctant to route sensitive operational data through a network co-owned by their largest rival, even with strong technical privacy controls in place. Onboarding costs and integration friction added to the strain.
What should shippers use now that TradeLens is gone?
Build visibility on direct integrations and open standards rather than a single platform. Set up API connections with your carriers and forwarders so shipment events flow into your own systems, and favour vendors that implement open, machine-readable data specifications. That way your visibility does not collapse if any one provider fails, and you can change suppliers without losing your data trail.