How AI is Transforming Freight Forwarding

Freight forwarding teams don't need another vague prediction that AI will “change everything.” They need to know what is already happening, where the numbers are credible, and which improvements show up in the margin.

The latest AI in freight forwarding statistics point to a clear pattern: adoption is rising quickly, but fully integrated automation is still uncommon. Forwarders are using AI first for document processing, quoting, forecasting, shipment visibility, and repetitive customer communication. The firms seeing real returns are not replacing their operators. They're removing the manual work that keeps experienced people away from exceptions, customers, and profitable decisions.

What do the latest AI freight forwarding statistics show?

The strongest evidence comes from several industry surveys published between late 2025 and 2026. They don't measure exactly the same population, so the numbers shouldn't be combined into one adoption rate. Together, though, they show where the market is heading.

A 2026 Boston Consulting Group and Alpega survey of more than 180 logistics providers and shippers found that:

• More than 40% of shippers now consider a logistics provider's AI capability when choosing a partner.
• About 40% of logistics service providers have moved beyond the pilot stage.
• Only about 10% have scaled AI across core operations.
• Nearly 80% of respondents see cost reduction and operational efficiency as the main reasons to invest in AI.
• Around half expect AI to cause some form of workforce transformation.

A separate 2025 Descartes transportation benchmark survey found that only 3% of respondents were not using AI in some form. The leading use cases were data-entry automation and the conversion of unstructured information, followed by route or load optimisation, load matching, and freight forecasting.

The gap matters. Many companies have an AI tool. Far fewer have an AI-enabled workflow.

The takeaway is simple: AI adoption starts with clerical work, not autonomous freight procurement. That is where the data is most structured, the risks are easier to control, and the payback is easier to prove.

Where is AI creating the most value for freight forwarders?

1. Quote preparation and pricing

Quoting is one of the best early AI use cases because the work is repetitive but commercially important.

A forwarder may need to extract:

• Origin and destination
• Incoterm
• Cargo type
• Weight and dimensions
• Container or equipment requirement
• Dangerous goods details
• Required delivery date
• Customs and local delivery requirements

That information often arrives across several emails, spreadsheets, PDFs, and photographs. An AI system can turn those inputs into structured shipment data, identify missing fields, and prepare a quote request or draft offer.

The value is not just speed. It is consistency.

Manual quoting creates several common margin problems:

• A surcharge is missed.
• A minimum charge is applied incorrectly.
• A buy rate is mistaken for a sell rate.
• A quote is sent with the wrong validity date.
• An air shipment is priced on actual weight instead of chargeable weight.
• A local delivery or customs fee is left outside the customer-facing total.

AI can flag these issues before the quote leaves the office. It still needs approval for unusual cargo, volatile capacity, or high-value customers. But a human should be checking the commercial decision, not retyping the shipment details.

This is where an integrated freight forwarding pricing workflow becomes more useful than a collection of disconnected AI tools. If the system reads the email but still requires an operator to re-enter everything into a TMS, much of the benefit disappears.

2. Document processing and compliance checks

Freight forwarding runs on documents, and documents are rarely clean.

Commercial invoices, packing lists, bills of lading, arrival notices, certificates, customs forms, and carrier statements often use different layouts and terminology. AI-powered optical character recognition and language models can extract the relevant fields, compare documents, and identify discrepancies.

The most practical checks include:

• Invoice quantity versus packing-list quantity
• Gross weight versus declared weight
• Container number format
• Missing consignee or notify-party details
• Inconsistent product descriptions
• Missing harmonised tariff information
• Dangerous goods declarations
• Shipment dates that conflict with the booking
• Duplicate documents

This does not remove the need for customs expertise. It makes the first review faster and gives the operator a ranked exception list.

The International Air Transport Association's 2026 air cargo technology report rated artificial intelligence and advanced analytics as very high-impact technologies for cargo operations. It also identified API technology as a major enabler, because AI cannot do much with data trapped in systems that do not connect.

That point is often missed. Better models do not fix broken data flows.

3. Shipment tracking and customer communication

Tracking is still one of the most repetitive tasks in forwarding. Customers ask where the shipment is, whether the vessel departed, whether the flight arrived, and whether customs released the cargo. Operators then search carrier portals, email overseas agents, and update internal systems.

AI can help by:

• Reading carrier status messages
• Normalising different milestone names
• Detecting late or missing events
• Predicting likely delivery delays
• Drafting customer updates
• Escalating exceptions to the right operator
• Answering routine status questions through a customer portal

The best systems don't simply show a container location. They explain what the event means operationally.

For example, “vessel departed” is not the same as “shipment is on schedule.” The AI should compare the actual departure against the booking, transhipment plan, terminal cut-off, and final delivery appointment.

Visibility also supports better internal prioritisation. A forwarder can focus on shipments with a missed connection, rolled booking, customs hold, or narrow delivery window instead of treating every open file as equally urgent.

4. Forecasting and capacity planning

AI is useful when freight demand changes faster than a team can update its spreadsheets.

Forecasting models can analyse:

• Historical shipment volumes
• Seasonality
• Customer order patterns
• Port congestion
• Blank sailings
• Fuel and currency movements
• Carrier capacity
• Trade policy changes
• Weather and disruption data

For forwarders, forecasting can improve decisions about block space, trucking capacity, warehouse labour, and customer pricing.

The Transporeon Transportation Pulse Report 2026 found that 44% of shippers were already using AI in transportation planning and optimisation. Freight procurement and real-time visibility followed, at 37% and 32%.

But forecasting is only as good as the underlying shipment history. If customer records use inconsistent lane names, missing weights, or unreliable milestone timestamps, the model will produce confident-looking nonsense.

That is why data cleaning should come before an AI forecasting project. Not after.

How is AI changing ocean, air, and road forwarding?

The use cases overlap across modes, but the commercial impact differs.

Ocean freight

Ocean forwarders can use AI to monitor:

• Vessel schedule reliability
• Port congestion
• Transhipment risk
• Container availability
• Demurrage and detention exposure
• Rolled bookings
• Free-time expiry
• Empty equipment constraints

AI is particularly valuable for exception management. A team should know which containers are most likely to incur storage charges or miss a delivery appointment, rather than receiving a generic list of every container in transit.

Global maritime trade remains enormous. UN Trade and Development data shows that 12.1 billion metric tons of goods were loaded for international maritime trade in 2024. At that scale, even small improvements in milestone accuracy and exception handling can produce meaningful operational savings.

Air freight

Air freight benefits from AI in booking selection, chargeable-weight calculation, capacity monitoring, and shipment prioritisation.

AI can compare service options based on:

• Chargeable weight
• Flight schedule
• Connection risk
• Transit time
• Product restrictions
• Temperature requirements
• Carrier allotments
• Expected handling delays

Air cargo is also becoming more tied to AI-related trade. IATA reported that air cargo carried more than two-thirds of the value of AI-related goods in 2025. AI-related consignments grew 20% year over year, while those goods represented 53.5% of the value of air-transported trade but only 7% of its volume.

That value density makes accurate pricing and exception management especially important. A missed connection on servers, memory chips, or data-centre equipment can cost far more than the freight charge alone.

The key point is the mismatch between value and volume. AI-related freight is not necessarily large in physical size, but it is commercially sensitive and often time-critical.

Road freight

Road forwarding teams can apply AI to:

• Carrier matching
• ETA prediction
• Route planning
• Tender acceptance
• Appointment scheduling
• Proof-of-delivery processing
• Freight invoice auditing
• Claims triage

The main benefit is faster coordination across a fragmented carrier network. AI can compare a load's requirements against carrier availability, equipment, service history, insurance, and geographic fit.

However, road freight also carries fraud and identity risks. AI should support carrier verification, not make an unchecked dispatch decision. New carriers, unusual payment requests, mismatched contact details, and sudden changes to banking information should trigger human review.

What is stopping freight forwarders from scaling AI?

The biggest barrier is usually not the model. It is the operating environment around it.

Poor data quality

More than half of shippers and carriers in the 2026 Transporeon report cited poor or inconsistent data as a major barrier to AI success.

Common examples include:

• Different names for the same port
• Missing container events
• Unstructured customer references
• Incorrect units of measure
• Duplicate shipments
• Old rate sheets still used in live workflows
• Manual status updates with no timestamp

AI magnifies bad process design. If a forwarder has no clear owner for a shipment milestone, the system will not solve the accountability problem.

Disconnected systems

A standalone chatbot may draft a response, but it cannot reliably update the quote, shipment file, accounting record, and customer portal.

That is why integration matters more than flashy features. APIs, webhooks, shared data models, and clear permissions are the infrastructure behind useful AI.

Unclear ROI

BCG's 2026 research found that only about one in ten logistics service providers reported measurable financial impact from AI. That does not mean AI fails. It means many projects are launched without a baseline.

Before implementation, measure:

• Average quote turnaround time
• Cost per shipment file
• Manual touches per shipment
• Tracking emails per operator
• Invoice error rate
• Gross margin leakage
• Time spent on document review
• Exception resolution time

Then measure the same figures after deployment. If the project cannot show improvement in one of those areas, it is probably a technology demonstration, not an operating investment.

What should a freight forwarder automate first?

Start with workflows that are frequent, rules-based, and easy to review.

A sensible sequence is:

1. Extract shipment data from emails and attachments.
2. Validate required fields and document consistency.
3. Draft rate requests and customer quotes.
4. Automate routine milestone updates.
5. Prioritise exceptions and delayed shipments.
6. Add pricing recommendations after the data is reliable.
7. Introduce controlled automation for bookings and carrier communication.

Don't begin with autonomous pricing across every trade lane. That is how margin loss gets hidden behind impressive dashboards.

Set approval thresholds. For example, an AI system may prepare a quote automatically when the lane, cargo, and service terms match known rules. A human should approve quotes involving dangerous goods, unusual dimensions, high-value cargo, weak historical data, or a margin below the company's minimum.

AI should handle volume. People should handle judgement.

Will AI replace freight forwarders?

No, but it will reduce the value of manual coordination.

The forwarders most exposed are those whose service consists mainly of copying rates, forwarding status emails, and rekeying documents. Those tasks are increasingly easy to automate.

The stronger role is advisory and operational:

• Choosing the right mode and service
• Managing exceptions
• Explaining trade-offs to customers
• Negotiating with carriers and agents
• Handling customs and compliance risk
• Designing resilient routing
• Protecting margin when conditions change

The IATA 2026 technology research frames AI as part of a wider technology stack that includes analytics, APIs, computer vision, and digital process automation. That is a more realistic view than the idea of one chatbot replacing an entire forwarding operation.

AI will change the team structure. It won't eliminate the need for people who understand cargo, capacity, documents, customers, and consequences.

FAQ

How is AI used in freight forwarding?

AI is used for quote preparation, document extraction, shipment tracking, ETA prediction, freight forecasting, carrier matching, invoice auditing, customer communication, and exception management. Most forwarders start with repetitive administrative work before moving into pricing or booking decisions.

What percentage of freight forwarders use AI?

There is no single global adoption rate because surveys measure different groups and definitions of AI. Descartes' 2025 transportation benchmark found that only 3% of respondents were not using AI in some form. BCG's 2026 survey found that about 40% of logistics service providers had moved beyond pilots, while only around 10% had scaled AI across core operations.

How does AI improve freight forwarding quotes?

AI extracts shipment details from emails and attachments, checks for missing information, applies pricing rules, compares service options, and prepares a draft quote. It can reduce turnaround time and prevent missed surcharges, incorrect chargeable weights, and inconsistent margin decisions.

Can AI predict freight rates?

AI can forecast freight-rate movements by analysing historical rates, capacity, demand, port conditions, fuel costs, seasonality, and market events. It should be used as a decision-support tool, not as a guarantee. Volatile events such as strikes, conflicts, regulatory changes, and sudden blank sailings can still defeat a forecast.

What are the biggest risks of AI in freight forwarding?

The main risks are poor data, incorrect document interpretation, overconfident recommendations, privacy issues, integration failures, and uncontrolled automated actions. Human approval is needed for high-value cargo, dangerous goods, customs decisions, unusual routing, low-margin quotes, and new carrier relationships.

How can small freight forwarders start using AI?

Small forwarders should begin with one measurable workflow, such as email-to-quote data extraction, document checking, or routine shipment updates. Establish a baseline first, connect the tool to the systems already in use, and track turnaround time, manual touches, errors, and margin leakage.

Is AI more useful for ocean, air, or road freight?

AI is useful across all three modes. Ocean forwarding benefits from schedule and demurrage-risk monitoring. Air forwarding benefits from chargeable-weight checks, capacity planning, and service selection. Road forwarding benefits from carrier matching, ETA prediction, appointment management, and invoice auditing.

What is the best way to measure AI ROI in freight forwarding?

Measure the workflow before and after implementation. Useful metrics include quote response time, cost per shipment, manual touches, document errors, tracking workload, exception resolution time, and gross margin leakage. Revenue growth alone is too broad to prove that an AI project worked.

AI is already transforming freight forwarding, but the statistics show a more practical story than the hype suggests. Adoption is widespread at the tool level. Scaled automation is still rare. The competitive advantage will go to forwarders that connect AI to clean data, real workflows, and disciplined human approval.

Last updated: September 2026