What a No-Substitution Clause Should Say
Most broker-shipper agreements are silent on who carries the loss when a load moves on an unvetted carrier.
The gap most agreements leave open
This is the clause we sign. Read any broker-shipper agreement for the word "substitute": in most of them it is not there, and neither is any language deciding who carries the loss when freight moves on a carrier nobody vetted. Ours says it plainly, and this briefing is the reasoning behind the language.
That silence is not usually malice. It is inheritance. Most broker agreements descend from templates written when the failure mode people worried about was a late truck, not an impersonated one. The clauses are thorough on liability limits, indemnity and payment, and quiet on the question of whether the entity that picked up the load is the entity named on the rate confirmation.
The consequence shows up exactly once, and only when something has already gone wrong.
What a working clause has to do
Four things, and the order matters because each one closes a hole the previous one opens.
Prohibit substitution outright, not "without consent".
A consent carve-out sounds reasonable and is where the clause leaks. Consent is verbal at 4pm on a Friday, unrecorded, and unprovable afterwards. If consent is permitted at all, require it in writing before the load is dispatched, and say that an unwritten consent is void.
Name re-brokering and co-brokering separately.
They are different acts and a carrier will argue the distinction. Re-brokering is tendering your load to another carrier. Co-brokering is tendering it to another broker, who then tenders it onward. A clause that prohibits only the first leaves the second available.
Put the liability back on the party that substituted.
This is the clause that matters and the one most often missing. Without it, a broker whose carrier re-brokered your load can point at a limitation of liability written for ordinary cargo loss and cap its exposure at a number set for a different kind of failure.
Make the bill of lading the evidentiary test.
A third name on the BOL is the cleanest proof a load was substituted. Say so in the contract: the carrier named on the rate confirmation must appear as the hauling party on the bill of lading, and any variance is a breach, not a discrepancy.
Where the weak versions leak
1. The consent carve-out
"Carrier shall not re-broker without Broker's prior consent" is the most common formulation and the weakest. It converts a prohibition into a permission structure, and permission structures are administered by whoever is under pressure at the time. Require it in writing, pre-dispatch, or do not permit it.
2. Liability that is capped by the wrong clause
Most agreements contain a general limitation of liability. If the substitution clause does not carve itself out of that cap, a total loss from an unvetted carrier is settled against a number written for a damaged pallet. State plainly that the limitation does not apply to losses arising from unauthorized substitution.
3. Silence on the downstream carrier
If a load is re-brokered and the downstream carrier disappears with it, the question becomes who had a duty toward you. A clause that says only "Carrier shall not re-broker" gives you a breach claim against a party who is frequently uncollectable. Add that the broker remains responsible for the acts of any carrier that moves the freight, however it came to be on the trailer.
Why this is not hypothetical
Fictitious pickups, where someone poses as the carrier you booked, are a substitution problem before they are a theft problem. That is what a re-brokering clause has to reach.
Below is the shape of a clause that does all four jobs. It is offered as a starting point to review with your counsel, not as legal advice, and the bracketed terms are the ones that have to be negotiated, not copied.
No substitution of carrier
Broker shall arrange transportation of Shipper's freight solely with motor carriers that Broker has qualified under its written carrier qualification standard. Broker shall not re-broker, co-broker, subcontract, interline or otherwise transfer any shipment to another broker, freight forwarder or motor carrier without Shipper's prior written consent obtained before dispatch. Verbal or after-the-fact consent shall be of no effect. The motor carrier identified on the rate confirmation shall appear as the carrier on the bill of lading, and any variance shall constitute a material breach. Broker shall remain responsible for loss, damage, delay and third-party liability arising from any shipment moved by a carrier other than the qualified carrier identified on the rate confirmation, and any limitation of liability elsewhere in this agreement shall not apply to such loss. Broker shall retain and produce on request its qualification and verification records for the carrier used on any shipment for a period of [retention period].
What to expect when you send it
A brokerage with a documented process will sign something close to this, possibly after negotiating the retention period and the consent mechanics. A brokerage without one will push back on the last sentence first, because producing verification records is only possible if they were created.
That reaction is itself useful information, and it arrives before you have tendered a load, not after you have lost one. We sign this clause as written, and on loads that move on our own fleet the question never arises: there is no substitution when the truck is ours.
FMCSA, Broker and Carrier Fraud and Identity Theft
Ask for artifacts, not descriptions
Every question below has a right answer, and every right answer is a document. That is the whole design. A brokerage that has the material sends it in a day. One that does not sends a paragraph about its commitment to safety.
None of this requires you to know how a brokerage works internally. It requires you to notice the difference between a process and an intention.
The written carrier qualification standard, as a document.
Not a summary, not a page on their site. The internal standard itself. If it does not exist as a document, that is the answer to the question, and it is worth knowing on day zero, not in year two when a claim asks who qualified the carrier.
What is verified at onboarding, and what is re-verified at tender.
Authority and insurance can lapse at any time, so a check done at onboarding can be months out of date. We re-check at every tender. The distinction between these two lists is the single best indicator of whether vetting is a process or a form somebody filled out once.
How the certificate of insurance is obtained.
The right answer is from the producer or agent named on the policy, with the broker named as certificate holder so cancellation notice reaches them. The wrong answer is that the carrier emails it. FMCSA states publicly that certificates presented to brokers can be fraudulent, so a PDF is a claim about coverage, not proof of it.
The disqualification criteria, and the override rule.
Every brokerage makes exceptions when a market tightens. The question is not whether exceptions happen. It is whether an exception is made by a named person, against stated criteria, and written down at the time. Ask what the last override was.
The double-brokering and fictitious-pickup controls.
How do they confirm the truck at your dock belongs to the carrier on the rate confirmation? Look for mechanisms: driver name and tractor and trailer numbers sent before pickup, a call to a number the broker sourced, not one the carrier supplied, a photograph at the gate. "We only use trusted carriers" is not a mechanism.
Retention: how long, in what system, who can pull it.
Cargo and liability claims arrive well after delivery. A process that cannot be reproduced in year three is not much of a process. Ask them to pull the file for a load they ran last quarter while you are on the call.
The insurance requirements in their carrier packet, and their own coverage.
Market convention is $1,000,000 auto liability and $100,000 cargo. Understand what those are: contract terms, not regulation. FMCSA sets no cargo minimum for general freight. Then ask the harder question, which is their contingent cargo limit and what their broker liability policy responds to.
Who is accountable for your freight, and is that the person who would be deposed? "Your account team" is not a name. On a small desk this question is easy. On a large one the answer tells you how many hands your load passes through.
Two questions that sound useful and are not
"How many carriers are in your network?"
Every broker sources from the same carrier pool. A large number describes a database, and a small number describes a narrow one, which may be exactly what you want. The count is unfalsifiable and uninformative in both directions. Ask instead how many carriers ran your specific lane for them in the last ninety days.
"Do you have capacity in my lane?"
The answer is always yes and it is nearly always true, because capacity exists in almost every lane at some price. It is not a lie, it is just not an answer. Ask what they would do on a Tuesday in peak when the first two carriers reject.
Why the file matters to you
Courts now allow shippers to sue brokers over careless carrier selection, which is why we keep verification records you can request. In a proceeding where your company is often a co-defendant, the quality of that file affects your exposure, not only theirs.
How to score the answers
Not by whether they say yes. By how long it takes to produce the document, and whether the document matches what they described on the call. A brokerage that has run this process before recognises the questions and has the material to hand. One that has not will treat the request as unusual, which is itself the finding.
FMCSA, Insurance Filing Requirements
Routing guides fail slowly, then all at once
Nobody discovers a broken routing guide on the day it breaks. They discover it the week spot rates cross contract on a lane that used to cover at primary, and by then the degradation has been running for a quarter.
The useful thing about that lag is that it is measurable. Every signal below sits in data you already have, and none of them requires a market subscription to read.
Six signals, in the order they appear
Primary tender acceptance is drifting down, not falling.
A carrier who accepted 95% of your tenders last quarter and accepts 88% this quarter has not exited your lane. They have started choosing which of your loads to take, which means your freight has become the marginal option in their network, not the base. That drift is the earliest signal and the one most often filed as noise.
Acceptance is holding but the accept is getting later.
Time-to-accept is a better instrument than acceptance rate because it moves first. A carrier deciding at hour six instead of hour one is shopping your load against alternatives before committing. Same acceptance number, different relationship.
Rejections are clustering by day of week.
Random rejection is capacity noise. Rejection concentrated on Thursday and Friday pickups means your loads are losing to something specific, usually a shipper whose freight positions the truck better for the weekend reset. That is a lane-design problem, not a rate problem, and paying more will not fix it.
You are going deeper into the routing guide on the same lanes.
Track the average tender position that covers, not just whether it covered. Moving from primary to second on a lane, consistently, is the guide telling you the primary rate is no longer market before any index does.
Backup carriers are becoming primary in practice.
If the carrier you designated third is covering more of a lane than the one you designated first, the guide on paper no longer describes the guide in operation. Every subsequent rate conversation is then being had against the wrong baseline.
Your own accessorial spend is rising on unchanged freight.
Detention, TONU and layover creeping up on lanes whose physical characteristics have not changed usually means carriers are accepting the load and then absorbing a worse experience than they priced. That is a relationship running down, and it precedes an exit.
What none of these tell you
They do not tell you whether your rate is right, because they cannot. Tender behavior is a signal about your freight's position in a carrier's network, and position is a function of the rate, the lane, the appointment rules, the dwell and the reload opportunity together. A guide can degrade with the rate untouched because a receiver started running two hours behind.
That is why the fix for a degrading guide is often not a rate change. It is finding out which of the other four variables moved.
Reading it against your own operation
Before repricing anything, check the three things a carrier prices that most shippers do not track.
The question a carrier is asking
How long is my truck sitting, and is it predictable enough to plan the next load around?
If I am ninety minutes late, do I lose the door or do I get worked in?
Where does this leave me, and is there freight there, or am I deadheading out?
Is my exposure capped, and does the cap get honoured without an argument?
The 48-hour move
If the signals are clear and you need coverage before you can run a proper bid, a mini-bid on the affected lanes is the instrument. It is not a compressed annual bid and running it like one is how it fails.
We wrote that up separately, because it is a different job with different mechanics.
FMCSA, Hours of Service of Drivers
BTS, Freight Transportation Services Index
The claim is true. That is the problem.
When a broker tells you they have capacity in your lane, they are almost certainly not lying. Brokers do not own trucks. They source from the same pool of for-hire carriers every other broker sources from, and in that pool there is capacity in nearly every lane at some price.
So the statement is true, unfalsifiable and uninformative, which is a difficult combination to interrogate on a sales call. The way through it is to stop testing the claim and start testing the mechanism behind it.
Four questions that cannot be answered with a reassurance
How many distinct carriers moved this lane for you in the last ninety days?
A number, not a network size. One carrier ninety times is a relationship. Ninety carriers once each is a load board. Both are legitimate, and they behave completely differently when a market tightens: the first has a truck for you and a reason to send it, the second has neither.
What happens after the first two rejections on a Tuesday in peak?
This is the real question and most answers to it are vague. Look for a sequence: who gets called, in what order, whether the rate moves and who authorizes that, at what point you are told. A brokerage that has done this has a procedure. One that has not describes an attitude.
What did you do the last time you failed to cover a load on this lane?
Every broker has failed to cover. The answer reveals two things: whether they will admit it, and whether the failure produced a change. "That has not happened" is either untrue or means they have not run enough of your lane to know.
Coverage risk is manageable if it arrives with lead time and unmanageable if it arrives at the appointment. Ask what their internal trigger is for telling you a load is at risk, in hours before pickup, and who makes that call.
Why "dedicated carrier network" does not survive contact
Shippers report disbelieving this phrase, and the disbelief is well founded. Unless a brokerage has contracted committed capacity, with volume guarantees running in both directions, the carriers it uses are also available to its competitors. Describing that as dedicated is a claim about intimacy, not exclusivity.
There is a version of the underlying point that is defensible and specific: a broker whose accounts are concentrated in a corridor gives its carriers repeat freight on the same lanes, which earns first call instead of a bid. That is a real mechanism and it is testable with question one.
What a smaller account list buys
Not more carriers. Fewer competing internal claims on the same trucks. When a market tightens, a desk with four hundred accounts is allocating capacity across four hundred sets of lanes. That allocation is where your freight either has a position or does not, and it is invisible from the outside.
Testing it with freight, not questions
The cheapest test is a small, real tender on a lane you care about, at a rate you consider fair, with a tight but honest transit requirement. Not a trap load and not your worst freight. What you are measuring is not whether they cover it, which they probably will, but four things around the covering.
How fast the quote comes back, and whether it asks the questions a competent desk asks.
Whether the rate confirmation names the carrier, the equipment and the terms without being chased.
Whether the driver, tractor and trailer numbers arrive before pickup or after you ask.
What happens on the one thing that goes slightly wrong, because something always does.
That sequence tells you more than a capacity conversation ever will, and it costs one load.