Spot freight happens: How machine learning is protecting shippers' budgets

05/11/2023 | 4 min

AI and machine learning can help shippers use data analytics to automate pricing and tendering decisions in the often volatile spot freight market.

The freight market is a fickle place, and even the best-laid plans can fall apart with a turn of the market. 2020 and 2021 have been prime examples of the fiscal havoc volatility can cause. Failing routing guides and rising tender rejections have forced shippers to either pay up in the spot market or let freight sit. Left at the mercy of carriers and increasing spot rates, we saw the trickle of overblown budgets in shrinking profit margins and inflated prices for goods.

Even in good times, most shippers avoid the spot market due to its tendency to destroy budgets. They view rigorous contracting as the best way to limit exposure, though in reality, most annual RFPs fail to produce rates that will still be relevant 12-15 months out. Without processes in place to dynamically adapt, navigating the spot market is an inevitability for which all shippers must prepare.

Thankfully, new technologies have evolved to solve this. This article examines the current spot freight market, challenges with annual budgeting models, and how new approaches to spot procurement help protect freight budgets using machine learning and data science.

Understanding the current spot market

Many interpret today’s sky-high spot freight costs as a consequence of a “capacity crunch.” In fact, overall tonnage moving in North America has declined since 2019. Capacity issues can instead be linked to issues that the pandemic may have triggered but will take years to unwind.

Changes in demand patterns: Rather than a spike in demand, there has been a shift in demand away from primary materials (lumber, steel, petroleum) to finished goods and consumer products. The consumer goods industry is generally more fragmented and less efficient for trucking, with more empty miles and long waiting times at distribution centres, creating the illusion of tightened capacity.

Workforce demographics: Concern about an ageing workforce plagues the trucking industry, as issues around low wages, excessive regulations, long work hours, and training limitations have made it difficult to recruit younger drivers.

Equipment and new capacity: Semiconductor shortages, plant closures, and steel availability are factors that currently impact the production of new tractors and trailers. Without an adequate supply of new equipment to replace outdated equipment as it is pulled off the road, overall capacity will continue to be strained.

The limitations of annual truckload budgeting

Traditional truckload budgeting assumes that annual RFPs can provide an accurate pulse on the market and assurance that contracted carriers would move the majority of freight in the next year. Management teams will make a probabilistic model of the likelihood of primary, secondary and tertiary carriers covering loads at agreed-upon rates. A small amount, perhaps 5-15%, is predicted to go to the spot market.

Problems arise when using miscalibrated models to gauge the market. Rarely do contracted rates remain relevant 12 to 15 months from when they’re first negotiated, and underestimating the amount of spot freight to be tendered puts shippers at an operational disadvantage that escalates quickly into a fiscal one. Overloaded planners take more shortcuts, and pricing is less rigorous. As freight falls out of contract, teams must make quick decisions on offers, often without a complete picture of the current market or fair rates.

To avoid overpaying on annual contracts, shippers are starting to look at shorter-term, mini-bid contracts and utilise technology to enable pricing agility. AI and machine learning platforms like Transporeon provide an opportunity for shippers to retake control of a negative situation by using data analytics to automate pricing and tendering decisions for spot freight – reducing the operational burden and ensuring scientifically optimised rates.

How AI and machine learning keep budgets in line

The Transporeon platform utilises AI and machine learning to create comprehensive carrier profiles based on truck positioning, reactions to previous offers, and third-party market data to predict carrier-specific pricing. The result is smart tendering, a shipper-driven model that pushes offers to carriers rather than requesting bids and automatically modifies pricing in response to market changes.

In this context, think of AI as a tireless negotiator with limitless attention to detail that spot-buys on a shipper’s behalf. AI can consider all the factors that a shipper is unable or unwilling to look up to ensure the best possible outcome for spot freight rates.

It is important to remember that Transporeon’s platform is not a magic wand that can revert spot market rates to fall in line with a shipper’s contracted annual budget. It is more akin to the airbag in a vehicle — while you hope you don’t need it, it offers the assurance of the best possible recovery in a bad situation. And as more freight is pushed into the spot market in today’s volatile truckload environment, having the security of a data-driven, automated negotiator protects a budget from being destroyed by overpriced spot rates.

Contact Transporeon to schedule a demo and learn more about how an approach to spot freight procurement that is driven by AI and machine learning can be your strongest strategy.

Ready to learn more? Schedule a demo today.