By Luis Lopez, AI transportation consultant, CEO of Go Hub.io Holdings Corp and subsidiaries, and host of the Freight Guru Podcast
A freight RFP takes weeks of work. Shippers clean lane data, invite carriers, compare bids and award lanes. Then the real test begins. An award is a commitment on paper, and the only way to know whether it holds up is to watch what happens when loads are actually tendered.
Contract carrier compliance is the practice of measuring whether awarded carriers accept the freight they agreed to haul, at the rates they agreed to, on the lanes they won. AI does not change the basic idea, but it makes the measurement faster, more complete and easier to act on. This article covers what to track, how AI helps, and where shippers need to be careful before drawing conclusions.
Why Awards Drift After the RFP
Awards drift for ordinary reasons. Market conditions shift, a carrier’s network changes, a lane turns out to be less balanced than it looked, or a carrier simply bid aggressively to win business. None of this is unusual. The problem is that drift often goes unnoticed for months because tender data lives in one place, spot purchases in another, and invoices in a third.
If you are still building your bid process, our guide to running a freight RFP covers the setup. This article picks up after the award.
The Core Metric: Tender Acceptance by Lane vs Award
The starting point is simple. For each lane, compare what each carrier was awarded with what it actually accepted. That means tracking:
- How many loads were tendered to the primary carrier on each lane.
- How many the carrier accepted, rejected or let expire.
- How the accepted volume compares with the volume the carrier committed to in its bid.
- Where rejected loads went next: a backup carrier, the routing guide, or the spot market.
Doing this by hand across hundreds of lanes is tedious. AI and analytics tools can pull tender records, normalize lane names that are spelled differently in different systems, and produce a lane-by-lane view in minutes. The value is not the math, which is basic, but the consistency and the speed of getting a reliable picture every week.
Detecting Rejection Patterns
Raw acceptance numbers hide patterns. AI is useful for surfacing them. Common ones include:
- Peak-period rejections. A carrier accepts reliably most of the year, then rejects heavily around holidays or end-of-quarter surges when spot rates are more attractive.
- Day-of-week patterns. Rejections cluster on certain pickup days, which may point to the carrier’s own network balance.
- Short lead-time rejections. A carrier rejects tenders sent with little notice but accepts those sent further ahead.
- Selective acceptance. A carrier accepts the easy loads on a lane and rejects the ones with difficult appointments or long dwell.
Each of these tells a different story, and each calls for a different conversation. Our article on what tender rejections tell shippers explores how to read them.
Measuring Spot-Rate Leakage
When a primary carrier rejects and backups do too, the load often falls to the spot market. The difference between the contract rate and what the shipper actually paid is leakage. Over a year, it can quietly erase the savings an RFP was supposed to deliver.
AI tools help by linking each spot purchase back to the original tender and lane, so the shipper can see which awards are leaking and by how much. Without that link, spot spend looks like a general cost of doing business rather than a symptom of specific awards that are not holding. For background on how contract and spot pricing interact, see our shipper guide to contract vs spot rates.
Building Carrier Scorecards
Scorecards turn compliance data into something carriers and procurement teams can discuss. A practical scorecard usually includes:
- Tender acceptance against award, by lane.
- On-time pickup and delivery.
- Tracking and communication quality.
- Billing accuracy and dispute frequency.
- Response time on tenders.
AI can generate these scorecards automatically and write a short plain-language summary of what changed since the last review. That summary should be checked by a person before it goes to a carrier, because a wrong number in a scorecard damages trust quickly.
Having the Conversation With Carriers
Compliance data is most useful as the start of a conversation, not a verdict. As an AI transportation consultant, the first thing I look at before any carrier review is whether the data shows a pattern the carrier can actually fix. A carrier rejecting short-notice tenders may simply need more lead time. A carrier struggling at one facility may be dealing with long dwell that the shipper controls.
Good conversations tend to cover:
- What the award said and what the tender history shows.
- Which specific lanes or periods are the issue.
- What the carrier sees on its side, such as appointment problems or network changes.
- What both sides will change, and when they will review it again.
The Data You Need
AI cannot measure what was never recorded. To track compliance well, shippers need:
- Tender records with timestamps, lane, carrier and requested pickup date.
- Carrier responses with accept, reject or expire status and the time of response.
- The award file with committed volumes and rates per lane and carrier.
- Actual spend by load, including spot purchases, linked to the original tender.
- Service data such as appointment times, arrival times and dwell.
Inconsistent lane naming and missing timestamps are the most common gaps. Cleaning those up is usually more valuable than any new tool.
Cautions Before Blaming the Carrier
Compliance numbers can mislead when shipper behavior is the real cause. Before concluding that a carrier is not honoring its award, check for:
- Short lead times. Tenders sent the same day or with minimal notice are harder for any carrier to cover.
- Volume far above forecast. If the shipper tenders much more than the RFP projected, rejections are a predictable result.
- Facility problems. Long detention, restrictive appointment windows or slow loading make freight less attractive.
- Tender errors. Wrong equipment type, incorrect addresses or duplicate tenders can show up as rejections.
Shippers who address their own side often see acceptance improve without any change in rates. Our piece on becoming a shipper of choice covers those habits.
Putting It Into Practice
Start with a weekly lane-level report of acceptance against award, add spot leakage, and review the top problem lanes with the carriers involved. Let AI handle the data gathering and pattern detection, and keep people in charge of judgment and relationships. An award only has value if it holds up in daily tendering, and measurement is how you find out.
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.
