
You awarded the annual bid in March. It's late November, two of your Texas lanes have been covering out of the guide since September, and the carrier who took Allentown to Charlotte stopped accepting in the third week of October. The next network RFP doesn't open until February and nobody has ten weeks to run one.
So you run a mini-bid: a short rebid of a defined set of lanes, worked off the lane file and carrier list you already built for the annual event. Emerge puts the turnaround at days rather than the two or three months a network bid takes. The mechanics are the easy part. What decides whether it was worth running is which lanes go into it, and most lane lists get built from a dataset that can't answer that.
The conditions this year are not subtle. Uber Freight's monthly executive freight update of 11 November 2026 has first tender acceptance at 79 percent in October, against 90 to 94 percent across the three years before, per its Q3 report. DAT's dry van report of 16 November 2026 has spot linehaul up 35.8 percent year over year. In June, J.B. Hunt's Spencer Frazier said routing guides set early in the 2026 bid season were crumbling at an accelerated pace.
The variance sort picks the wrong twenty lanes
The standard selection method is a sort. Pull a market benchmark, set it against your awarded linehaul, rebid everything more than 10 or 15 percent under. Emerge's trigger list starts at roughly a 15 percent spot-to-contract spread, and plenty of teams stop there. The list is defensible. It also ranks lanes by how wrong the rate looks rather than by what the wrongness costs.
Take two lanes both 18 percent under market. The first moves 40 loads a week and the primary still accepts 88 percent of them, because it's a backhaul that fits their network. The second moves four a week, rejects nearly all of them, and each one gets bought on the spot board at whatever Tuesday costs. Variance scores them identically. One is taking money out of your budget this quarter and it's the small one.
Rank by cover cost instead: rejected loads in the period, times the gap between the award and what you actually paid to move them, annualised. A lane that fails four times a week at $900 over beats a lane that fails twice a month at $400 over, and no benchmark variance surfaces that.
What the invoice knows that the rate table doesn't
The gap between award and cover rate is the visible half. The rest sits in accessorial lines, which you see only by reading covered loads as a cohort.
A backup carrier picking up at two days' notice bills differently from a primary that's run the lane fifty times. They don't know which gate at the DC takes live unloads, so they sit, and detention lands. Nobody told them the receiver uses a lumper service, so the driver pays and it returns as a pass-through with a receipt attached. Dispatch at 6pm, miss the window, and you add a TONU or a redelivery.
The pattern we see most often is accessorial spend per load on covered freight running several times what the primary bills on the same lane, with detention the largest single code. That belongs in the cover cost. Leave it out and the business case understates the problem.
So the ranking runs on automated freight invoice audit data, not TMS rate data. The TMS holds what you agreed; the invoice holds what happened. Plants taking inbound raw material see it first, one receiving window driving detention across every carrier on the lane, which is where freight audit for manufacturers usually starts.
Normalise fuel before you rank anything
One thing quietly ruins this arithmetic. The award defers fuel to a contract matrix, the spot cover came in all-in, and subtracting one from the other measures a surcharge, not a rate gap. Fuel is four separate calculations wearing one line item, and mixing two of them puts a healthy lane at the top of your list.
Rejections that aren't about your rate
Before a lane goes on the list, look at how its rejections are distributed. Several patterns recur and only one is about price.
Rejections clustered at a single origin are a dwell problem. That carrier has put a number on four hours in your yard, and a rebid buys you the same number from somebody new. Fix the appointment process or write detention terms with teeth, then reprice.
When they cluster on short lead time, your tender timing is the variable. A Thursday 4pm tender for a Friday 6am pickup gets declined at any rate that isn't a spot rate. Move to 48 hours and watch the lane for a month first.
One carrier rejecting while the other three keep accepting is that carrier's network changing. Reallocate down the guide and move on.
Then there's the one nobody wants to write down. If you committed 40 loads a week in the annual bid and tendered 12, the rejections are a memory. Carriers track awarded against tendered volume and will price an under-tendered lane with a premium on it, fairly enough. Go in with the real forecast and say so.
And a lane that keeps failing after a clean rebid was never a pricing problem. Check dwell and appointment compliance before you buy a third rate on it.
Put your overpriced lanes in the same event
Mini-bids run as a one-way ratchet almost everywhere. You pull only the lanes where the carrier is unhappy, every lane reprices upward, and the result gets written up as a save because somebody measured against spot cover instead of last year's award. Do that twice a year and your contract book tracks the market up and never follows it back down.
Run the variance sort in both directions and put a handful of over-market lanes in the same package. Some are over for a reason you'd defend: the lane was hard to cover last spring and you bought certainty. Others are over because a carrier's pricing analyst was sharper than yours at 11pm on bid night and nobody has looked since.
Price the risk first, because putting a lane in play tells an incumbent accepting at 96 percent that it's available. We'd still do it where you're clearly over and three carriers want the freight. We wouldn't touch a lane where the incumbent is the only carrier who'll run it. Taking 4 percent off a lane and losing the one truck that covers it reliably is a trade most shippers regret inside a quarter.
A mini-bid calendar that fits in ten days
Pull the lane list Monday from tender logs and audit data. Send the package Tuesday to incumbents plus two or three carriers per lane who aren't on it. Bids due Friday noon, award Monday, routing guide updated in the TMS Tuesday and live the following Monday. A twenty-lane event is a few hundred emails and calls to carriers who each care about two lines in your file, which is most of what freight rate negotiation actually is.
The step teams skip is the TMS update. An award that lives in a spreadsheet and never reaches the guide keeps tendering to the old primary at the old rate, and somebody asks three weeks later why the lane is still rejecting. Then do one thing by hand: tell an incumbent they lost a lane, instead of letting them work it out when the tenders stop. You'll want their pricing desk in February.
Run the sort this week
Take the last 90 days: rejected loads per lane, what you actually paid to cover each one, accessorial spend on covered loads against the primary's, and the lanes sitting above market. Sort by annualised cover cost and see where the list falls off.
Fewer than ten lanes clearing the bar means you don't have a mini-bid. You have four routing guide edits and one awkward call with an account manager. Forty clearing it means the annual event is the problem, and the routing guide failure you're reading in the invoice started at award time.
Book a walkthrough to see Miles and Penny build a cover-cost ranking from your own tender logs and paid invoices, then turn the top twenty lines into a bid package.
Frequently asked questions
- What is a freight mini-bid?
- A mini-bid is a short rebid of a defined set of lanes, run between annual RFPs using the lane file, pricing template and carrier list from the last full event. Because the templates and carrier contacts carry across, it closes in days rather than the two or three months a network bid takes. Shippers use one when part of a routing guide stops performing but the rest of the award is still sound.
- How do you decide which lanes go into a mini-bid?
- Most teams sort by variance to a market benchmark and rebid everything more than 10 to 15 percent under. That ranks lanes by how wrong the rate looks rather than by what it costs you. Rank instead by annualised cover cost: rejected loads in the period times the gap between the award and what you actually paid to move them, with accessorial spend on the covered loads included. A low-volume lane that fails every week often outranks a high-volume lane that sits further under market but keeps accepting.
- Does a high tender rejection rate on a lane always mean the rate is too low?
- No. Rejections clustered at one origin point at dwell, and rebidding a dwell problem buys the same detention exposure from a new carrier. Rejections clustered on short lead time point at tender timing. Rejections concentrated in a single carrier while the others on the lane keep accepting usually mean that carrier's network changed, which calls for a routing guide reallocation rather than a bid.
- How many lanes should a mini-bid cover?
- There's no fixed number, but the size tells you whether a mini-bid is the right tool. Fewer than about ten lanes is usually better handled as routing guide edits and a direct conversation with the incumbent. Forty or more suggests the annual award itself was mispriced, and a targeted rebid won't fix that.
See it on your freight
Stop Overpaying On Freight Invoices
Penny checks every invoice against your contracts and what the shipment actually did, then disputes what doesn’t hold up. You pay 30% of what she saves you, and nothing if she saves nothing.
Found this useful? Share it with your team.
Share on LinkedIn