Skip to content

Bid and RFP analysis

Bid too high and you lose the contract, bid too low and you lose margin. Know which lanes to win, and at what price, before you commit.

  • Free 30-minute call
  • No commitment
  • You own the code
Typical industries
Trucking and fleet, Logistics and transport
Problem type
Pricing and range
You receive
A documented model, run on your data

In short: A freight bid analysis checks every lane in a customer’s tender before a carrier bids. A lane is a regular route between two places. It shows which lanes fit the routes and depots the carrier already runs. It also shows what each lane costs and which price wins the contract and still earns enough.

The problem

Three customers want your rates for next year’s transport by Friday. You take the average cost per kilometre and add the usual margin, your profit. Then you send the rate sheets back. You win most of the routes you bid on. Six months later the trucks on those routes come home empty. Nobody can say which contract is losing money.

Signs your bids leak margin

A customer asks you and other carriers for prices with a request for proposal (RFP), a formal tender. You price each RFP on its own. A lane is a regular route between two places, say Basel to Milan. Every lane gets the same blended rate, one average price per mile. That rate rests on averages: average miles, how full a truck runs on average and average fuel. You win where your price is lowest. Those turn out to be the lanes where you earn least. Trucks run empty. And you pay drivers for trips to loads that never cover their cost.

Why lane-by-lane bidding fails

A freight bid fixes your price for a contract period, often a year or more. Win the wrong lane at the wrong rate, and no dashboard turns red. The damage spreads thinly across thousands of shipments. Your trucks deliver far from their home depot. There is no load for the trip back, so they return empty.

The usual approach treats each RFP, and each lane in it, as its own decision. It prices every lane at cost plus a fixed margin, the profit you add on top. Or it tries to win the contract at any price and bids lower than everyone else. It rarely checks whether a lane fits the routes your trucks already run. And it rarely compares several RFPs with each other. You win the lanes where you earn least. And you promise lanes that need the same trucks at the same time. Each bid looked reasonable on its own. Together they lose margin across all your contracts.

What changes

Today, you send one blended rate per lane and hope the mix works out. With our model, every lane in every open tender gets three scores. One for cost, one for how well it fits the routes your trucks already run, and one for your chance of winning. You get a ranked list: bid here at this price, decline there. You review it, add what you know about each customer and send the bids.

Freight bid analysis optimisation: what you get

You know what each lane costs you before you price it. You see which lanes fill trucks that would otherwise run back empty, and which only tie up trucks you need elsewhere. For each bid, you get the price that earns the most. Your chance of winning is already counted in. So you stop giving money away on bids you’d have won at a higher price. Some lanes send your trucks far from your depot, with no load for the trip back, so they return empty. Those lanes are marked to decline.

We build the model on your own lane costs, your terminals (the depots your trucks run from) and the lanes you already run. It also uses your win rate, the share of past bids you won. We test it against bids you won and lost in the past, before it goes live. If you bid at cost plus a fixed margin, you don’t see whether a contract fits the lanes your trucks already run. The model shows it for each lane.

  • A model written around your own bid and RFP analysis rules.
  • Results measured in the numbers you already track.
  • It slots into how you already work, and you can check every answer it gives.

How it works

Every lane in your open tenders goes into one model. For each lane, the model weighs the cost, your chance of winning it and how well it fits where your trucks already run. How many trucks and drivers you have sets the limits. You get a ranked list: where to bid, at what price and which lanes to decline. You add what you know about each customer and send the bids.

Lane profitability analysis

The first piece is lane profitability: do you earn money on this lane or not? Most carriers don’t know, because they price every lane with one blended rate. Margin is made or lost lane by lane, though. A bid based on the fleet average can look fine on paper and still lose money on every load. And because it’s the cheapest bid, it’s the one you win.

A lane’s full cost includes the empty miles back. It also includes the deadhead, the empty drive to the pickup. It also includes driver pay, fuel and tolls. And it includes the cost of getting the right truck to the place where it loads. Detention, the paid waiting time at a customer’s dock, counts too. Once you know what each lane costs, you can choose which ones to bid.

Driving-time rules decide what a lane costs

A driver day is one day of work for one driver. How many driver days a lane needs depends on how long your drivers may sit at the wheel. Daily driving has a cap, and only a few longer days are allowed each week. Drivers need a break after a set time at the wheel and a minimum rest each day. Those rules turn hours on the road into driver days, and driver days into cost.

Take two lanes. On the first, the round trip fits inside one driver’s daily driving limit, if loading and waiting stay short. On the second, it runs an hour over. That costs one of the few longer days a week, or a night away from base. On a rate sheet, the two lanes cost the same per kilometre. But the second needs more driver days and so costs you more. The model sees that, because it counts driver days for each lane.

Network fit scoring

The second piece is network fit: how well a lane fits your terminals and the routes your trucks already run. Most bid analysis stops before this question. A lane can earn little on its own and still be one of your best decisions. That happens when it fills a backhaul, a return trip you were running empty. A lane that looks fine alone can also leave a truck far from your nearest terminal. Tomorrow’s planned loads then have no truck.

Network fit scoring checks every lane in the tender with three questions. Where do your trucks already run? Where are your terminals? Where are trucks and drivers already short? Supply chain network design asks where to put sites and trucks. Network fit bid analysis asks the opposite: does a new lane fit the trucks and drivers you already have?

Win probability and bid pricing

The third piece is the price itself. The best price is the highest rate that still wins. The higher your price, the lower your chance of winning. A win-probability model calculates that curve for each lane and each customer. Margin is what a lane earns after its costs. Expected margin is that margin times the chance of winning the lane. Say a lane earns 1,000 francs of margin at a 40% chance: that’s 400 francs. The model finds the price where expected margin peaks. Bid lower, and your chance of winning rises. But you give away margin for a chance you didn’t need.

Volume commitments add a second question. RFPs often ask you to promise a minimum number of loads. Fixed costs are what you pay however many loads run, such as the truck lease. A lane with few loads may never cover them. A low-margin lane with a large volume commitment ties up trucks you need elsewhere. So RFP response optimisation checks two things for each lane. Do the loads cover the fixed costs? And will the trucks be missing elsewhere?

Joint optimisation across the bid set

Joint optimisation decides on all lanes from all open RFPs in one calculation. It decides together which lanes to bid on, at what price and for how many loads. The goal is the highest expected margin. The model is a mixed-integer linear program (MILP). That’s maths for plans in which some choices are yes or no: bid on this lane or not. Thousands of those choices depend on each other: a truck you promise to one lane is missing on another.

The model never plans more trucks or drivers than you have. And it only picks lanes that one of your terminals can serve. For each lane, it knows how your chance of winning falls as the price rises. You get a ranked list of bids to pursue, the price for each and the lanes to decline. The lanes you win this season decide where your trucks run next year. That’s why it pays to choose them together.

Common questions.

Something else? Ask us directly.

Work out the lane with its own costs, never with the average of your whole fleet. A lane is a regular route between two places. Count the kilometres with a load and the empty kilometres back or to the next pickup. Add driver days, tolls and waiting time while loading and unloading. Last, check how the lane fits your other routes. If it brings a load for a return trip that would otherwise be empty, you earn extra. If it ends far from your depot with no load for the trip back, the truck returns empty, and that costs you.

Group similar lanes together. On lanes in the same region, with the same truck type and similar customers, you win about equally often. So the model estimates their chance of winning together. Lost bids carry as much information as won ones. Record the rate you bid. Where you hear it, record the winning rate too. The estimate sharpens with each tender season.

Yes, when it fills a return trip you’re already running empty. The lane alone may earn little or even lose money. The round trip as a whole still earns more. Work out your profit across all lanes twice: once with this lane and once without it. If the profit is higher with the lane at the price you’d bid, the bid is worth it.

Sometimes it can, and if that’s the case, we’ll tell you. Standard software is built for an average operation, though. Yours runs on its own rules: your contracts, your exceptions, the things you know by heart. A custom model writes those rules in, and that’s where the better plan comes from.

You don’t need either to start. We build the model from the data you’ve already got, even if it lives in three spreadsheets and one person’s head. You get a system you run yourself. We train your team until they no longer need us.

You see a first result with the prototype, before you commit to anything bigger. We start with a free 30-minute call about your operation and the model that would fit it. Then we build a working prototype and run it on your numbers.

Tell us what you’re planning.

In a free 30-minute call with a specialist, you find out whether optimisation can help.

  • Free, no commitment
  • You talk to a specialist directly
  • A clear go or no-go