Connect with us

Non classé

Ocean Freight Procurement in 2026: A Research-Based Approach

Published

on

Ocean Freight Procurement in 2026: A Research-Based Approach

December 4, 2025

Blog

Forwarders and BCOs rely heavily on tendered contracts in order to plan pricing for global ocean freight movements. But what if large swaths of the tendered contracts are simply completely irrelevant? Banking on joint research with the MIT Center for Transportation, this article challenges the basic assumption of tenders, makes a case for more strategic use of the spot market, and forces a hard, data-backed look at actual performance.

There have been plenty of examples of volatility and disruption in freight markets over the last few years, from the pandemic and port strikes to the Red Sea crisis and trade wars.

These events have triggered demand swings, impacted operations and lead times, and often sent freight rates climbing or plunging – all of which affect shipper and forwarder decisions on how to allocate freight volumes between contracts and the spot market.

Across modes, having a freight contract in place does not always mean shipments will get moved. Dr. Angi Acocella, a Freight Lab researcher at MIT’s Center for Transportation and Logistics, conducted research on procurement decisions and performance in the FTL market, and, via a joint survey with Freightos, on container shipping procurement as well. These studies explored how shippers make tendering decisions and provide insights on how to improve strategic procurement decisions across modes to reduce risk and costs and improve reliability.

See Dr. Acocella’s presentation of the survey results here.

The studies show that:

Both road and ocean shippers rely heavily on contracts – with most shipping at least 70% of volumes by tender – and mostly treat spot as a backup option. They also show that a significant share of contracts – on average 70% of FTL contracts – go unused.

Unused lanes come with unnecessary costs, including a 7% year-on-year increase in contract rates, while the strategic use of spot for these types of lanes can mean significant savings in both time spent on wasted negotiations and in shipping rates.

Index-linked contracts have also been shown to increase revenue for carriers and provide better reliability and lower costs for shippers.

And digital tools that provide visibility of contract performance, market intelligence, and a dynamic channel to communicate and share information with LSPs are also key to optimizing the freight procurement process.

What follows are the key findings from these recent surveys and the best practices for freight procurement and tendering that emerge from the research.

Are All Ocean Tenders Actually Used? The Findings

The surveys showed that in the FTL market, most shippers moved 90% of their volumes by contract with the remaining 10% going by spot. Ocean shippers also relied heavily on long term contracts, with more than half moving 70% of volumes or more via contracts.

In both FTL and container, the spot market is mostly used as a backup to contracts. The move to spot is most often due to the uncertainty, unreliability or unavailability of the contracted capacity, or to unexpected swings in their own demand. Often these pushes to spot are triggered when spot prices become significantly out of sync with contract rates.

At the same time, 40% of shippers do report using the spot market as a first option on some lanes, usually for trades with infrequent, less predictable and lower volumes than contracted lanes. But this finding also means that 60% of ocean shippers rely on contracts even for low volume lanes.

Freight Lab research found a Pareto principle at work for FTL shippers: in general, about 20% of shipper lanes account for about 80% of their volumes, with a long tail 80% of lanes accounting for just 20% of – mostly low or inconsistent – volumes.

For many of those long tail lanes, however, volumes don’t materialize at all. Acocella refers to lanes with unused contracts as “ghost lanes,” and research shows that shippers underestimate how many contracts go unused.

A majority of both FTL and container shippers estimate a ghost rate of 25% or less. But FTL data suggests that on average, closer to 70% of contracted lanes go unused.

Ghost lanes come with added costs beyond the resources wasted negotiating unused contracts. MIT research shows that high rates of FTL ghost lanes one year resulted both in lower acceptance rates that year and in 7% higher contract costs – often at levels higher than spot rates – the following year as carriers factor in a premium to compensate for unreliability.

The Actionable Insights

These findings – the volatility in these markets, the heavy reliance on contracts which can often become unreliable, the use of spot mostly as a second-choice backup, and the share of contracts that go unused – begged the key research questions:

What is the optimal balance between freight contracts and spot shipments? What approach is most likely to maximize reliability and efficiency and minimize costs?

The resulting Freight Lab research produced the following key components to constructing an optimized contract/spot procurement portfolio*

*These results are based primarily on research for the FTL market, with applicability to ocean likely based on the above survey and with more research on this area underway now.

Learn from past performance: A look at contract portfolio and spot usage and performance – its mix, utilization and reliability by lane – from the previous year is critical to decision-making for the coming year. Shippers and forwarders should determine where contracts performed well, where they were underused (or not used at all), and price levels paid relative to the market.

Based on this picture, companies can determine where and with which LSPs to renew contracts on highly or regularly utilized lanes. For lanes where little or no volumes materialized shippers and forwarders should consider a strategic direct-to-spot approach, given the high costs of ghost lanes.

Respect the market cycle: Shippers and forwarders should also consider where the market is in its cycle, and how contracts performed in previous instances of this phase. In tight markets it generally does not benefit shippers to negotiate hard for discounted contract rates as – often regardless of a shipper’s history with a given carrier – contract price competitiveness becomes the carriers’ priority. In soft markets, shippers have more leverage and should contract with their most reliable carriers. But here too, low-ball rates can often mean poorer performance, especially if market rates increase.

Index link some contracts – Contracts linked to an index allow rates to fluctuate in some relationship to the spot market. Significant spot changes are a main driver of poor contract performance: when spot rates climb too high above a contract’s rate, carriers roll volumes or apply premiums. When spot rates fall too low, shippers no-show and shift to the spot market or renegotiate contract levels.

Allowing a contract to float along with an index removes this incentive to deviate from the contract. And though index linking exposes carriers and shippers to fluctuations in revenue or costs, this risk can (either be accepted as the cost of reliable service/volumes, or) be hedged through derivatives called Forward Freight Agreements, effectively locking in contracted service at a set cost or revenue level even while paid rate levels may change.

Learn more about index-linked contracts here.

Acocella advises shippers to pilot index linking with a carrier they trust and with whom they have an existing relationship, and on mid-volume lanes – especially where volumes may not be consistent throughout the year – to start. Once the concept is proven, this tool can be expanded to other lanes. Research in the FTL market showed that index-linked contracts, especially on these types of lanes, often resulted in carriers realizing higher revenue and shippers getting better reliability, and often lower costs than when using typical contracts or just spot.

Track performance – Shippers and forwarders should not only evaluate past performance during tendering season, but should monitor performance and utilization during the contract period too. Generating visibility of your contract (and spot) portfolio, and tracking where volumes are materializing and which contracts are being underused, will let shippers understand how carriers are performing and how they, the shippers, are performing as a partner. Understanding where you stand can help you adjust in real time rather than carry unnecessary costs or incur higher costs or poorer reliability down the road.

Leverage tech – New tech tools help evaluate past performance or track the current status of freight tenders, costs, volume (or lack thereof) by lane, and spot usage, all of which is crucial to optimizing freight portfolios.

These tools also represent the opportunity for a significant leap in freight procurement efficiency: Beyond digital tools that help shippers compare their contract portfolio to market prices and provide market intelligence to support decision making, logistics tech is also enabling more efficient tendering by creating a dynamic, digital channel to communicate, negotiate, share data, and even make spot or contracted bookings with LSPs.

Tech that simplifies or automates parts of the freight procurement process, or otherwise reduces the time spent on negotiations and procurement, is increasingly key to overall streamlined procurement.

The Bottom Line: Don’t Tender Everything

Recent research shows a heavy reliance on long-term contracts for both road and ocean freight shippers, with the spot market mostly reserved as a backup. But shippers also carry significant wasted costs through unused or underutilized contracts, with analysis showing better efficiency via the strategic use of spot for these lanes.

These studies suggest certain key procurement best practices for shippers, including investing in visibility of their contract/spot portfolio and performance; the strategic use of spot on low volume lanes; consideration of the current phase in the market cycle; implementation of index-linked contracts, and leveraging digital tools to provide that necessary visibility as well as automate or digitize time-consuming tasks like requesting tender offers, spot rates and even placing bookings.

Judah Levine

Head of Research, Freightos Group

Judah is an experienced market research manager, using data-driven analytics to deliver market-based insights. Judah produces the Freightos Group’s FBX Weekly Freight Update and other research on what’s happening in the industry from shipper behaviors to the latest in logistics technology and digitization.

Put the Data in Data-Backed Decision Making

Freightos Terminal helps tens of thousands of freight pros stay informed across all their ports and lanes

The post Ocean Freight Procurement in 2026: A Research-Based Approach appeared first on Freightos.

Continue Reading

Non classé

Warehouse Orchestration: Solving the Daily Breakdown Between Plan and Execution

Published

on

By

Warehouse Orchestration: Solving The Daily Breakdown Between Plan And Execution

In most warehouses today, the problem is not whether work gets done; it is how much effort it takes to keep everything aligned and on track. Every day, there is a breakdown between the plan and executing the plan. Labor plans, inbound schedules, picking priorities, and automation all operate from valid assumptions, but not always the same ones. The gaps between them are filled in real time by supervisors and teams, making constant adjustments. That is what keeps operations running, but it is also what makes them fragile.

It is a challenge many operations recognize. Even with modern systems in place, execution still depends heavily on human coordination. Warehouse orchestration is the shift from managing tasks independently to coordinating the entire operation and ensuring decisions across the system stay aligned as conditions change. The best way to understand what that means in practice is not through a system diagram, but through the lens and experience of the people running the floor.

Consider Maria, a warehouse supervisor responsible for keeping a high-volume operation on track. She is experienced, practical, and steady under pressure, but what she is really managing is not just work; it is complexity.

At any given moment, she balances labor availability, work queues, inbound variability, equipment status, and shifting order priorities. Those inputs are not wrong. They are just not aligned. It is her job to bridge that gap in real time.

A shift that starts “normal” … until it does not

Maria arrives before the floor fully wakes up. Her first stop is not the dock or the pick module; it is yesterday’s reality. What shipped? What did not? Where did the backlog form? Which waves did not behave as the plan assumed? She is not looking for blame; she is looking for drift. Drift is what turns into firefighting later.

Demand shifted over the weekend, but the pick face still reflects last week’s reality. One area is short-staffed; another has idle labor. When the team built the labor plan, it made sense, but the day had already moved on. The team scheduled inbound; however, it is not predictable. Every ETA is a best guess, and how trailers show up rarely matches how they appear on a screen.

Individually, nothing here is catastrophic, but warehouses do not fail all at once. They gradually lose alignment between plan and execution. The team compensates in real time by moving people, reprioritizing work, working around automation delays, and making judgment calls. And the shift “works,” but there is a cost:

Overtime, which did not need to happen.

Detention fees, which show up later.

Service misses, driven by wrong priorities rather than a lack of effort.

Leaders who spend more time reacting than improving.

These challenges are the reality across many operations. Execution is strong, but coordination is fragile.

The real bottleneck: decisions are fragmented

Most warehouses are not short on tools. They have WMS, robotics systems, labor tools, and planning solutions. Each one does its job well, but they do not make decisions together. Each system optimizes its scope based on different priorities or timings. The gaps between them are filled manually by people like Maria. In an environment with less variability, that might work, but in most cases:

Demand changes faster and more frequently.

Labor is less predictable.

Automation introduces new dependencies.

Customer expectations continue to rise.

Under these conditions, static plans, especially labor plans and wave structures, can drift out of sync before the shift is halfway through. That is when the operation starts relying on “manual heroics.” Experienced supervisors keep things running. It is hard to scale, and even harder to sustain.

AI-driven warehouse orchestration: keeping the operation aligned

Warehouse orchestration and the power of AI address this gap. Because it is not just about executing tasks, it is about coordinating decisions across the operation and using intelligence to see, analyze, and recommend actions with full visibility to all the variables. Instead of managing isolated activities, intelligent orchestration continuously aligns:

Labor to demand.

Inbound and outbound priorities.

Work sequencing across zones.

Automation with human workflows.

It does this in real time, as conditions change. Variability is constant, and it is not realistic to eliminate. The goal is to see the risk earlier, respond faster and more consistently, and prevent disruption.

Back to Maria: when the system helps carry the load

Now imagine Maria running that same Monday, but operations now behave like a connected ecosystem, not a collection of islands. Before the shift even starts, she is not just reviewing what happened yesterday. She is looking at a forward-facing view that is already adjusting based on incoming signals. She is getting visibility into risk early before it is a problem. Inbound appointments are not just a schedule; they are a ranked set of trade-offs that balance urgency, detention risk, inventory needs, and outbound commitments. Her decisions are clearer because the system prioritizes them, reflecting business impact. Slotting does not rely on disruptive, periodic re-slot projects that leave the pick face to decay. Instead, optimization and learning continuously shape placement, folding the highest value moves into natural replenishment windows and explaining the “why” in business language.

And during the shift, when one area starts falling behind, Maria does not have to guess the best move. She can see the impact of her options:

Shifting labor.

Reprioritizing tasks.

Adjusting sequencing.

Instead of relying on instinct and experience alone, she has visibility into how decisions affect the entire operation. She is still in control, but the system is helping her avoid problems instead of chasing them. And that changes how the shift feels. It is not static; it is dynamic, but stable.

The key ingredients: unified data, SaaS, AI & ML, connected systems

Behind the scenes, this comes down to unified data, SaaS, AI, ML, and systems that work together. When you connect your warehouse systems, add real-time operational signals and visibility to systems outside of the warehouse, and apply AI and ML for speed and precision, you are working from a single source of truth and an interconnected ecosystem of systems. As a result, users make decisions with a broader context. Then the operation starts to learn; outcomes inform future decisions, improving how the system responds over time. And now, humans are not the only thing holding the performance together.

Why this matters right now

For supply chain leaders, this is not only about efficiency. It is about operating in a world where volatility is constant. Across industries, the specifics vary, but the challenges are consistent:

Handling demand swings without inflating labor costs

Scaling operations without scaling complexity

Maintaining service levels under pressure

The operations that succeed are the ones that do not just react faster; they are the ones that operate in alignment.

The shift ahead

A single, modern technology will not define the future of warehouse management. It will be defined by how well operations coordinate across people, systems, and workflows in real time. That is what intelligent warehouse orchestration enables. It turns the warehouse from a collection of well-run processes into a connected system that can adjust continuously. Because in the end, the goal is not just to execute the plan. It is to keep the plan from breaking when the shift starts.

By Tammy Kulesa
Senior Director, Solution & Industry Marketing, Blue Yonder

Tammy is the Senior Director of Solution and Industry Marketing, leading go-to-market strategy and thought leadership for Blue Yonder Cognitive Solutions for Execution, and the LSP Industry. With over 20 years of experience in technology marketing and nearly a decade focused on retail, logistics, and supply chain, Tammy brings a deep understanding of the operational and strategic challenges facing today’s supply chain leaders. A passionate advocate for innovation and collaboration, Tammy has a proven track record of connecting market needs with transformative solutions.

The post Warehouse Orchestration: Solving the Daily Breakdown Between Plan and Execution appeared first on Logistics Viewpoints.

Continue Reading

Non classé

How Operational AI Turns Supply Chain Recommendations into Action

Published

on

By

Supply chain AI cannot stop at better insight. To create operational value, AI recommendations must connect to workflows, execution systems, approval paths, and measurable outcomes.

Artificial intelligence is quickly becoming part of the supply chain technology conversation. Vendors are adding copilots, recommendation engines, autonomous agents, and predictive analytics to planning, transportation, warehousing, procurement, and visibility applications. The promise is clear: better decisions, faster responses, and more adaptive operations.

But there is a critical distinction that supply chain leaders need to keep in view. An AI system that identifies a problem is not the same as an AI system that helps solve it.

A demand-planning model may identify a likely stockout. A transportation model may flag a lane disruption. A supplier-risk model may detect a deteriorating delivery pattern. Those are useful insights. But unless the system can connect that insight to an action pathway, the burden still falls on the planner, transportation manager, procurement team, or customer service group to decide what happens next.

That is where many AI deployments will either create real value or stall out.

For a deeper look at the architecture behind operational AI, including A2A, MCP, RAG, Graph RAG, and connected decision systems, download the full white paper: AI in the Supply Chain: From Architecture to Execution.

Insight Is Not Execution

Supply chains do not run on insight alone. They run on orders, shipments, purchase orders, inventory moves, carrier tenders, production schedules, warehouse labor plans, customer commitments, and exception workflows.

A recommendation that remains in a dashboard is not yet operational AI. It is decision support. Decision support can be valuable, but it does not fundamentally change the operating model unless it becomes part of the execution process.

The question is not simply, “Can the AI make a recommendation?” The better question is, “Can the organization act on that recommendation in a controlled, auditable, and timely way?”

For example, if an AI system predicts that a regional distribution center will run short of inventory, several action pathways may be available. The company might expedite inbound supply, rebalance inventory from another facility, substitute a product, modify customer allocation rules, or adjust promised delivery dates.

Each action has a cost, a service implication, and a governance requirement.

Operational AI must understand those pathways. It must also know which actions it can recommend, which it can execute automatically, and which require human approval.

The Execution Layer Matters

This is why integration with core execution systems is so important. AI cannot operate effectively if it sits outside the systems where work is actually performed.

For supply chain AI to become operational, it must connect to transportation management systems, warehouse management systems, order management systems, ERP, procurement platforms, supplier portals, customer service workflows, and control tower environments.

Without these connections, AI may diagnose problems faster, but it will not necessarily resolve them faster.

The difference is material. An AI assistant that says, “This shipment is likely to miss its delivery appointment,” is useful. An AI-enabled workflow that identifies the delay, calculates downstream service risk, recommends a carrier alternative, checks cost thresholds, initiates an approval workflow, and updates customer service is much more powerful.

That is the move from analytics to operational intelligence.

Human-in-the-Loop Still Matters

This does not mean every AI recommendation should become an automated action. Supply chain decisions often involve tradeoffs among cost, service, risk, inventory, and customer relationships. Many require judgment.

The more practical model is tiered autonomy.

Low-risk, high-frequency actions may be automated. Moderate-risk decisions may require planner approval. High-impact exceptions may require escalation to a manager or executive.

This is not a weakness. It is a design requirement.

A well-architected operational AI system should know when to act, when to recommend, and when to escalate. It should also capture the outcome so the system can learn whether the decision improved performance.

Closed-Loop Learning Is the Real Prize

The most important capability may not be the first recommendation. It may be the feedback loop that follows.

Did the expedited shipment prevent the stockout? Did the alternate supplier meet the delivery date? Did the inventory transfer protect service without creating a shortage elsewhere? Did the customer accept the revised promise date?

These outcomes should not disappear into operational noise. They should feed back into the intelligence layer.

That is how AI becomes more than a static recommendation tool. It becomes a learning system embedded in the daily operating rhythm of the supply chain.

What This Means for Buyers

Supply chain leaders evaluating AI-enabled software should press vendors on action pathways. The relevant questions are straightforward.

Can the system connect recommendations to execution workflows? Can it distinguish between automated, approved, and escalated actions? Can it operate across functions, not just inside one application? Can it create an audit trail? Can it learn from outcomes?

The vendors that answer these questions well will move beyond AI features. They will become part of the operating architecture.

The next phase of supply chain AI will not be won by the tool that produces the most impressive recommendation. It will be won by the systems that help companies act faster, with more control, better context, and measurable outcomes.

The post How Operational AI Turns Supply Chain Recommendations into Action appeared first on Logistics Viewpoints.

Continue Reading

Non classé

test

Published

on

By

The post test appeared first on Logistics Viewpoints.

Continue Reading

Trending