How to Use Carrier Scorecards in Your Freight RFP

Connecting the two is the highest-value change you can make to a freight RFP, because unreliable capacity is a cost rather than a service footnote.

Why rate-only awards cost more than they save

A carrier bidding below the incumbent but accepting 70% of tenders does not deliver that rate. Rejected loads cascade to backup carriers and the spot market, and your real cost per load on that lane rises above both bids.

The scorecard is what tells you in advance which bids are credible. A carrier's history on your freight is the best available predictor of whether they will haul it at the price they quoted. For how to build one in the first place, see What Is a Carrier Scorecard.

What to bring into the bid

Tender acceptance, by lane. Network averages hide the lanes that matter. A carrier at 94% overall can sit at 71% into your hardest facility.

On-time pickup and delivery, by lane. Separate appointment performance from requested-date performance, since carriers often hit one and miss the other.

Contract compliance. Accepted loads against the volume the carrier committed to in the last award. This measures whether a carrier keeps the commitments they make in an RFP, which is exactly what you are asking them to make again.

Actual cost per load against the contracted rate. A carrier whose actual average has drifted above contract is telling you what the next year looks like.

Claims and dispute history where it is material to the freight.

Three ways scorecards change an award

As a qualification gate

Set a minimum performance level a carrier must clear to be eligible for specific lanes. This works best on freight where a service failure is expensive, such as production inbound or customer-critical delivery.

Make the threshold explicit to carriers before bidding. A gate applied silently after bids are in reads as arbitrary and damages the relationship you are trying to build.

As a scoring input alongside rate

Weight performance next to price rather than using it as a tiebreaker. A reasonable starting point for truckload is rate and historical performance at roughly equal weight, adjusted toward performance on lanes where failure costs more.

In GoodShip, bid comparison surfaces in-line context for each bid based on historical carrier performance, market benchmarks including booked and posted rates and ranges, and previously awarded lanes and rates, so the performance record sits next to the number during evaluation rather than in a separate file.

As a constraint on the award itself

Ranking carriers lane by lane is not the same as awarding a network. Once you add real constraints, the best-scoring carrier on a lane sometimes has to lose it so the network works.

GoodShip's AI Scenario Builder models award options under performance, asset, and incumbency rules, including minimum on-time pickup and delivery performance, required asset mix, and lane-level and network-level incumbency thresholds, and runs side-by-side scenarios so you can compare award structures before committing. There is a worked example in Freight Award Scenario Analysis.

Tell carriers where they stand before you ask them to bid

A carrier who sees their scorecard for the first time inside a bid event has no chance to fix anything, which makes the data a judgment rather than a tool.

In GoodShip, carriers get self-service access to their own scorecards through a dedicated carrier portal, where they can see current contracts, revenue to date, and performance against metrics like on-time delivery and tender acceptance. Scorecard visibility is controlled by the shipper, and carriers also receive a weekly scorecard email. Shifts in tender acceptance and routing surface as Detected Actions, service warning emails can be sent to carriers from the email library, and late load disputes are resolved in the platform, with each action logged and archived by lane.

Carriers that have been working from the same numbers all year arrive at the bid ready to address the gaps instead of disputing them.

Close the loop after the award

The scorecard only works as an RFP input if it keeps running afterward. Track acceptance, service, and actual cost per load against every awarded lane, and the next bid starts with evidence about how the last one performed.

When a lane degrades badly enough, that record is also what justifies pulling it into a mini-bid rather than waiting a year.