By Luis Lopez, AI transportation consultant, CEO of Go Hub.io Holdings Corp and subsidiaries, and host of the Freight Guru Podcast
Freight rate indexes show up everywhere: in trade press headlines, in RFP discussions, in carrier pricing meetings and in contract clauses. A shipper hears that “the index is down” and wonders why their rates are not. A carrier sees an index rise and expects higher pay on every lane. Both reactions come from treating an index as a precise price instead of what it really is: a statistical summary built from a particular set of data, with particular blind spots. This article explains how spot and contract benchmarks are built and how to use them well.
What a freight rate index is
A freight rate index is a number, or a series of numbers over time, that summarizes the price of moving freight in some segment of the market. It might cover dry van truckload nationally, reefer in a region, a specific origin-destination lane, LTL, intermodal, or ocean container routes. Some indexes report an average rate per mile or per container. Others report a value relative to a base period, so the number shows change rather than an actual price.
Every index makes choices about what data to include, how to clean it and how to weight it. Those choices determine what the index can and cannot tell you.
Spot vs contract benchmarks
Spot benchmarks
Spot benchmarks measure the price of one-off loads booked close to the ship date, usually through brokers or load boards. Spot prices react quickly to changes in capacity and demand, so spot indexes move faster and swing wider than contract indexes.
Contract benchmarks
Contract benchmarks measure the rates shippers and carriers agree to in advance, typically through annual or periodic bids. Contract rates are stickier. They often reflect market conditions from the time the bid was run, not today. For a deeper look at the two pricing models, see contract vs spot freight rates.
The gap between spot and contract benchmarks is itself a signal. When spot sits well below contract, carriers have trouble filling trucks and shippers gain leverage at the next bid. When spot rises toward or above contract, carriers start rejecting contract tenders and shippers lose leverage. That relationship is one of the indicators discussed in the freight market cycle explained.
Where the data comes from
Posted rates
Some data comes from rates posted or offered on load boards and marketplaces. Posted rates are plentiful and timely, but a posted rate is an asking price. It may not be what the load actually moved for, and loads that sell quickly may never be posted at all.
Paid rates
Other data comes from actual transactions: invoices, freight payment records, or rate confirmations submitted by brokers, carriers or shippers. Paid rates are closer to the truth, but they arrive with a delay and depend on who chooses to contribute data.
Survey and reported data
Some benchmarks are built from surveys of market participants or from rates that companies voluntarily report. These can cover segments where transaction data is thin, but they rely on the accuracy and representativeness of the respondents.
The limits every index carries
Sample bias
No index sees the whole market. A benchmark built from broker transactions underrepresents freight that moves directly between shippers and asset carriers. A benchmark built from large shippers’ payment data may not reflect small shippers. If your freight profile differs from the sample, the index may not describe your market.
Lag
Paid-rate data takes time to collect. Contract rates reflect agreements made months earlier. Even a weekly index can describe conditions that have already shifted. Treat any index as a view of the recent past.
Averaging hides lane detail
A national average blends headhaul lanes with backhaul lanes, short hauls with long hauls, and easy freight with difficult freight. A lane leaving a strong outbound market may price very differently from the national figure. The dynamics of backhaul vs headhaul lane balance are a big reason averages mislead.
Fuel treatment
Some indexes include fuel, some exclude it, and some report both. Comparing an all-in rate to a linehaul-only benchmark produces a false gap. Always confirm which one you are looking at.
Mix shifts
If the mix of lanes or lengths of haul in the sample changes from one period to the next, the average can move even when no individual lane’s price changed. Good index providers adjust for this, but not all do, and the method is not always disclosed.
How shippers should use indexes
- Use them for direction, not for exact price. An index is good at telling you whether the market is loosening or tightening. It is weak at telling you what your specific lane should cost.
- Compare against your own history. Your own paid rates on your own lanes are the best benchmark you have. Use the index to explain changes in that history, not to replace it.
- Match the segment. Use a reefer benchmark for reefer freight, a regional benchmark for regional lanes, and a lane-level benchmark where one exists with adequate data.
- Read the methodology. Know whether the index uses posted or paid rates, includes fuel, and how it handles small samples.
- Watch the spread. The relationship between spot and contract benchmarks often matters more than either one alone.
- Be careful with index-linked contracts. If a contract adjusts rates to an index, define exactly which index, which segment, the calculation, the review frequency, and what happens if the index is discontinued or changes methodology.
How carriers and brokers can use them
Carriers can use spot and contract benchmarks to judge whether a lane is worth committing capacity to at a bid, and to recognize when the spot market is likely to pay better than a contract offer. Brokers can use them to set expectations with shippers and to sanity-check quotes. Neither should treat a benchmark as a floor or ceiling. Real lanes carry their own costs, from deadhead to dwell time, that no average captures.
Where AI fits
As an AI transportation consultant, the first thing I look at when a team wants AI-driven rate benchmarks is the data underneath. Models can blend public indexes with a company’s own paid-rate history to produce lane-level estimates, and they can flag when a quote sits far outside the expected range. That approach is covered in AI lane rate benchmarking with historical data. But a model trained on biased or lagging data inherits that bias and lag. It can also look more precise than it is. Ask how any estimate was built and how wide its uncertainty is before relying on it in a negotiation.
The short version
- Spot indexes move fast and reflect short-term capacity; contract indexes move slowly and reflect earlier bids.
- Posted rates are timely but are asking prices; paid rates are accurate but delayed.
- Every index has sample bias, lag and averaging effects.
- Use indexes for direction and context, and your own lane data for price.
For more conversations on freight and AI, subscribe to the Freight Guru Podcast.
About the author: Luis Lopez is a Miami-based AI transportation consultant and logistics entrepreneur, the CEO of Go Hub.io Holdings Corp and subsidiaries, and host of the Freight Guru Podcast.