Part 1 of a two-part series on how AI is reshaping freight procurement.
The freight market has changed. Procurement hasn’t kept up.
Freight procurement has followed the same basic rhythm for decades. Run an annual or semi-annual RFP. Lock in rates. Execute against those contracts. Repeat.
That model made sense when the freight market was relatively stable. It no longer is.
Here is what is happening right now: Truckload capacity is tightening. Smaller carriers are being forced out by rising insurance costs and tightening operating margins. Over 330,000 drivers in the United States face regulatory exit due to CDL and English language proficiency enforcement.
The carrier pool is not just volatile. It is structurally shrinking.
When capacity tightens, basic economics take over. Fewer available trucks means spot rates increase. As spot rates rise, carriers start rejecting contracted loads because they can earn more in the open market. Tender rejection rates climb. Routing guides that looked solid on paper begin to fail.
Tender rejection rates have surged past 14%, a level not seen since the post-COVID disruptions of 2022. At 14%, roughly one in seven loads is being rejected. That is not a rounding error. That is a systemic problem.
And here is the part that matters most for procurement teams: your contracts were negotiated on carrier-provided data. The carrier performance assumptions are just that: they’re frozen in amber and may reflect conditions that may no longer exist.
The market will not slow down to match manual processes.
Why traditional transportation management systems fall short on procurement
Most transportation management systems were designed for execution: plan loads, tender shipments, manage invoices. They were built as transactional systems, many of them 20 or 30 years ago, and then migrated to the cloud without fundamentally changing what they were designed to do.
When it comes to procurement, these systems rely on static inputs. Annual bid data. Fixed routing guides. Historical carrier relationships. Manual benchmarking.
That creates a structural gap. By the time teams revisit contracts, market conditions have already moved. Procurement teams discover problems only after they show up in the real world: rising spot market exposure, deteriorating carrier performance, increasing transportation costs that nobody flagged until the quarterly review.
In other words, freight procurement inside most legacy systems is reactive by design. It responds to problems after they occur rather than detecting shifts as they happen.
For supply chain leaders evaluating their technology stack, this is the question to ask: does your system tell you when a lane is drifting above market, or do you find out when market conditions force a choice?
What proactive freight procurement actually looks like
The shift happening now in freight and logistics is not incremental. It is a model change.
Event-based procurement treats sourcing as something that happens on a schedule: annual bids, quarterly reviews, manually triggered mini-bids when something breaks. Proactive procurement treats sourcing as an ongoing operating discipline where AI-driven systems detect opportunities and take action as market conditions change.
That means three things happen differently.
Detection happens automatically. Instead of waiting for a quarterly review to surface a problem, the system identifies lanes where contracted rates have drifted above market benchmarks, where carrier performance is deteriorating, where contracts are approaching expiration, or where lanes are running without contracted coverage. These signals are monitored in real time using both live market data and verified carrier performance, not carrier-reported claims.
Decisions are informed by real data, not stale benchmarks. Traditional procurement decisions are often based on data that is months old by the time it informs action. Proactive freight procurement uses real-time transportation visibility and market intelligence to benchmark rates and score carriers against what is actually happening in the network right now. The foundation matters: the quality and source of your data determines whether your procurement decisions reflect reality or approximate it.
Execution is embedded, not bolted on. When the system identifies an opportunity, it does not just generate a report and wait for someone to act on it. It recommends or launches targeted digital mini-bids, engages carriers, manages negotiation rounds, and surfaces ready-to-approve contracts. The entire workflow,from signal to sourcing action to contract,happens within the system.
This is what an AI-native transportation management system makes possible. Not AI bolted onto a legacy architecture. Intelligence embedded into how the system operates from the ground up.
How the AI Freight Procurement Agent works inside Intelligent TMS
project44’s AI Freight Procurement Agent is the operational mechanism behind this shift. It sits within Freight Procurement Analytics and operates as a standalone module or as part of project44’s Intelligent TMS.
The Agent works on a simple framework: Detect, Decide, Act.
Detect. The Agent benchmarks contracted rates against current market conditions and evaluates carrier performance by lane. It identifies where you are overpaying, where performance is degrading, where contracts are expiring, and where you have uncovered lanes with rising spot exposure. It does this using project44’s logistics data graph, which processes over 1.5 billion shipments annually across more than 259,000 connected carriers in 186 countries. The Agent then layers in market data for decisioning. Critically, this is built on observed execution data, not self-reported carrier claims.
Decide. When conditions shift, the Agent scores carriers based on configurable criteria: cost, service, reliability, and tracking quality. It evaluates both your existing carrier relationships and carriers from across the network that run your lanes well but that you may not have contracted. It uses live market context and your own historical booking data to determine the best path forward.
Act. The Agent launches targeted mini-bids autonomously, manages negotiation rounds with carriers, and presents negotiated contracts for human approval. No contract is accepted without a person confirming it. Every action is fully logged and auditable. Your team sets the guardrails, rate thresholds, carrier eligibility, contract parameters, negotiation policies, and the Agent operates within them.
You do not need to replace your TMS to get started
One of the most common objections to modernizing freight procurement is the assumption that it requires replacing your existing transportation management system.
It does not.
project44’s AI Freight Procurement Agent is available as a standalone capability. It works alongside your current TMS by ingesting your data, applying network-scale intelligence, and pushing recommended changes back to your system of record. No rip-and-replace migration. No 18-month implementation timeline.
This modular approach is central to how project44’s Intelligent TMS is designed. Start where it matters most and expand across modes, regions, and business units over time.
The organizations that move now to lock in capacity, protect margins, and modernize procurement will have a structural advantage. The ones that wait will continue buying freight reactively, at whatever rate the market dictates, whenever the routing guide breaks.
Speed is leverage. And the market is not waiting.
Webinar
Watch a live demo of the AI Freight Procurement Agent as it identifies opportunities, engages carriers, and surfaces contracts for approval.
