A Freight Tender That Still Works in Month Six

A freight tender is a single day's price. It then has to last a year.

On the day, everything looks settled: the lanes are listed, the bids are in, the lowest sensible price wins. Then the market moves, your volumes do not arrive in the shape you promised, and by month six the carrier who won is turning down the jobs it bid for. The tender did not fail on the day. It failed because nothing in it was built to cope with what changes.

This article is about building that in. It is for the person who writes the tender, usually procurement, and it assumes you will be buying full truckloads on a set of lanes.

What a tender actually produces

A US study from MIT describes the structure well, and Thai buyers will recognise it. A tender does not give you one carrier. It gives you three layers:

• A primary carrier per lane, who has bid a price and is expected to take the work.
• A list of backup carriers, usually the bidders who lost, offered the work when the primary says no.
• The spot market, for what both refuse.

The paper's summary of earlier US work puts the split at about 72% of loads moved by the primary, 23% by backups and 5% on the spot market. Those are US numbers, not Thai ones. The shape is what to take from them: the price you won in the tender covers only the first layer, and the real cost of your year is the blend of all three.

A small example shows why. Say a primary carrier's price is 100, and every load it refuses goes to a backup at 118. The paper, citing earlier US work, says backup prices can reach more than 18% above the primary's contracted price depending on the market, so that is a fair test case. In its own data, the primary accepted 81.9% of loads in a soft market and 68.5% in the tight one that followed.

• At 81.9% acceptance, your blended price is 103.3. Your tender price is 100.
• At 68.5% acceptance, it is 105.7.

That is before any spot loads, which cost more. The tender price did not change. The price you paid did, and the only thing that moved was the market. So the first rule is: judge a tender by what the lane costs after refusals, not by the winning bid.

Why the price goes stale, in Thai numbers

You do not need US data to see the problem. Look at Thai diesel.

Retail diesel (B7) in Bangkok averaged 31.94 baht a litre every month from May to September 2025, then 31.31 in October. In February 2026 it was 29.94. In April 2026 it was 44.13, and it was still 39.25 in June. A tender awarded in October 2025 and fixed for twelve months was priced when diesel was 31.31. Six months later, in April 2026, diesel was 41% higher.

The Ministry of Commerce's road freight index shows the same turn. In the second quarter of 2025 it was up only 0.3% on a year earlier, which is a flat year. A year later it was up 11.1%, with the pickup truck class up 12.0% and the trailer class up 11.3%. The index itself gave no warning of what was coming.

So the tender has to say, in advance, what happens when the price moves. The mechanics for fuel are in fuel surcharges in Thailand: name the published series, the grade, the base price and date, and the way the surcharge steps. Write that clause into the tender document, not into a later negotiation. A carrier asked to hold a flat price across a swing like the one above will either price in a buffer at the start or come back in month six. The first costs you every month. The second costs you the relationship.

Fuel is not the only thing that moves. For the rest of the rate, name a published index and a rule. For example, the Ministry's index by vehicle class is published quarterly and is free to read. Decide in the tender what size of move reopens the non-fuel part of the rate, how often you look, and what happens if the index falls as well as rises. Keep it symmetrical, or the carriers who read it will price the risk in.

Your volume is a promise, and carriers read it carefully

The same MIT paper looked at what makes a primary carrier accept or refuse a load, and the answers are about you, not just about price.

• How regularly you send work. The paper measures, for each load, the share of the previous four weeks in which the shipper sent that carrier at least one load on that lane. On lanes where this was lower, asset-owning carriers needed a contract price about 7% higher in the soft market to keep acceptance at 90% (617 against 682 dollars a load in the paper's example, with the spot price held at 1,000), and about 2.5% higher in the tight one.
• How much volume swings week to week. For asset-owning carriers on lanes with high swings, the price needed for 90% acceptance was 17% higher than on steady lanes in the soft market, and 26% higher in the tight one.
• Volume above the awarded amount. Carriers report, the paper says, that they can usually find trucks when a week's loads are within about 10% of the award. Past 20% they often cannot. For loads more than 20% above the award, the price asset carriers needed to hold 90% acceptance rose 9% in the soft market and about 13% in the tight one, which brought it almost to the spot price.

These are measured on US loads. Thai lanes and Thai carriers will differ. But the mechanism is the same everywhere trucks are a fixed daily resource, and the first two are things you can fix before the tender.

What this means for how you write one:

1. Tender on the weekly shape of each lane, not the annual total. Give a typical week, the range, and the busiest week. How much capacity to commit shows how to build that number. A carrier that is told "2,600 loads a year" has to guess whether that is 50 a week or 200 in one month.
2. Price in bands. The Handbook of Logistics and Distribution Management describes a hybrid price: a unit price with a guaranteed volume, so that demand swings do not leave the carrier with idle trucks. For trucks, ask each bidder for three prices per lane: loads up to about 10% over the weekly award, loads between 10% and 20% over, and loads beyond that. You will pay for flexibility. The point is to know the price before you need it.
3. Keep thin lanes out of the contract. If a lane gets a load every few weeks, awarding it a twelve-month price buys you a refusal. Leave it to be bought load by load, and say so in the tender.

Name the backup, and its price, in the tender

The backup list is the layer buyers neglect, because nobody is paying for it. In the MIT paper it is a list that has no contract with the shipper, so its prices are not binding, and its carriers are only loosely expected to produce trucks.

So make it real. For each lane:

• Award a primary and a named second carrier, and ask both to price it.
• Write down the second carrier's price and how many loads a week it can take.
• Say what a refusal triggers. For example, a load refused by the primary goes to the second carrier at its tender price, not at whatever the day's quote is.

Doing this at tender costs you a little in negotiating leverage. It saves you far more than the surprise at the back end, because you have already seen what the second carrier's price is while carriers were competing for the work.

Ask the questions the carrier should have to answer

The Handbook is written for larger outsourcing tenders but two of its points carry over to a truckload one. First, give every bidder the same information, and if one bidder asks a question, circulate the answer to all of them. Second, ask for answers in a fixed format, so you can compare like with like. Among the items the Handbook says to request in the response are the procedure for price increases and the penalty for ending early.

For a truckload tender that gives you a response form with these lines, each of which should be filled in by every bidder:

• Price per trip on each lane, in the three volume bands
• The fuel mechanism the bidder proposes, with index, base and trigger
• How the non-fuel part of the rate can change, and on what evidence
• Maximum loads per week the bidder will commit on each lane
• Notice either side must give to end or change the award
• Payment terms the bidder assumes

That last line matters because it is where a carrier's real price hides. A bid that assumes 30 days from a carrier who is then paid in 75 is a different price from the one on the form.

The Handbook also describes an evergreen arrangement: a fixed price and performance targets agreed for twelve months, with no end date and a long notice period either side, so that nobody has to renegotiate from scratch every year. It is worth considering when you and the carrier both want to stay, as it turns the renewal from a cliff into a review.

The test: write the month-six answers before you award

Before you award, write down the answers to five questions. If you cannot, the tender is not finished.

1. What happens if diesel moves 15%? Who pays, from what date, against which published number?
2. What happens if our volume on this lane is 25% above the award for three weeks? Which price applies, and what can the carrier refuse?
3. What happens if the primary refuses a load? Who gets it, at what price, and who records it?
4. How will we know the award is going wrong? Count how often each primary accepts the load first time, lane by lane, from the first month. The carrier measures worth tracking shows how.
5. When do we look again? Put a review date in the tender at month three and at month six. It is far easier to adjust a clause on a date you agreed in advance than to ask for one when the market has already moved.

What to do before your next tender

• Pull your actual loads per week for each lane for the last six months. Mark the lanes that are thin or swing widely, and take them out of the award.
• Write the fuel clause and the non-fuel index rule into the tender document, with a base date, as part of the specification.
• Ask for three volume-band prices, and a named backup on each lane.
• Put the response in a fixed form so bids can be set side by side.
• Fix a month-three review date before the bids come in.